Prior to Birth The Tu~ita episodes in Indian Buddhist literature and art

Christian Luczanits1

Prior to birth, it is said, the Bodhisattva had stayed in Tu~ita heaven and descended from there to assume his last worldly existence. Indeed, the notion that the last birth of a Buddha follows a sojourn in Tu~ita heaven is found in all textual sources that also narrate the birth. However, there are considerable differences in these sources concerning the context of the descent from heaven and in the details narrated. In principal the events prior to birth become more and more elaborate and miraculous. This article focuses on the relationship of the visual evidence to textual sources. Although I begin with the texts, it should be clear that no preference should be given to any of the sources. Textual tradition(s) and visual tradition(s) not only are inherently distinct but also relate to each other in a variety of ways. These interrelations and transformations can only be understood when the wider context ofthe occurrence is taken into account as well. In the case of this study, the wider context considered is

1 This article is a revised version of a section on my MA thesis finished in Vienna, Austria, in 1993 under the direction of D.E. Klimburg-Salter and Ernst Steinkellner. German translations of the Lalitavistara have originally been done in the course of a seminar with Ernst Steinkellner. Maurizio Taddei worked extensively with me on the Tu~ita depictions in Gandharan art. Dieter Sch1ingloff kindly provided me with the analysis of the relevant sections in Chinese textual sources that have been prepared for him by Valentina Stache-Rosen between 1970 and 1980. Besides these Haiyan Hu-von Hintiber, le. Heesterman and Chlodwig Werba helped with different aspects concerning the MA thesis. Now, more than a decade later, the Lumbini International Research Institute headed by Christoph Ctippers not only gave the impetus to revise my previous study by organising an international seminar on the Buddha's birth (which I sadly could not attend then) but also enabled the revision for publication due to a research grant of which this is the first result. Finally, Elizabeth Rosen Stone gave me valuable comments and Harry Falk provided his interpretation of the problematic Bharhut inscription. Christian Luczanits the life of the Buddha and its placement within the relevant text or monument. Although practically all major primary sources on the life of the Buddha that have come down to us begin with the Buddha's descent from Tu~ita heaven, the secondary literature often ignores this event.2 The primary reason for this is the 'historical' approach that is favoured by western and western trained scholars and writers. Thereby plausible details, thus considered possible historical facts, are separated from legendary material to create a life ofthe 'historical Buddha'.3 This process very much reverses the exaltation process visible in the development in the story ofthe Buddha's life and, considering the conclusions ofBareau's study on the earliest sources on the Buddhas youth,4 appears equally arbitrary. The visual sources, too, may have had their part in the selection visible in secondary literature. As we will see, Tu~ita depictions are rare and often their identification is problematic and the descent itself is integrative part of queen Maya's dream. Generally, many of the aspects considered here have already been touched upon by Ernst Windisch in his detailed study on the birth of the Buddha.s He also notes, that the story of the Buddha's birth actually begins with the sojourn in Tu~ita heaven.

Textual Sources Among the relevant Indian textual sources the Lalitavistara (LV) is arguably the latest, reaching its final form in its extant Sanskrit and Tibetan versions as late as the 7th or 8th century (see below).6 The LV also contains the most elaborate account on the events prior to birth. The

2 This is not the case in accounts based on the Tibetan tradition where the miraculous events are integrative part of the tradition. See for example: Csoma de Koros 1839: 231-232; Schiefner 1849: 2; Koppen 1906: 74. 3 Examples are: Pischel191O: 15; Oldenberg 1921: 108; Lamotte 1988: 15-23. 4 Bareau 1974: 26~274.

S Windisch 1908. 6 An even later source on the Buddha's life, the Jinacarita, is not considered here, since it has no relevance for the interpretation of the visuals (see Nonnan 1983: 158).

42 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art following survey of the texts begins with those that are least detailed concerning the relevant events and thus show the greatest divergence to the LV. Given that this survey of the textual sources serves as comparison to visuals, special consideration is taken concerning the illustrative value of the narration. This value may essentially be of two kinds, sequential and in detail. Under sequential illustrative value I understand any information concerning temporal and/or spatial sequences as well as sequences of visually representable action. Dialogues, for example, may be essential in developing the story but are almost impossible to depict. lllustrative details, in contrast, are elements that can be depicted in certain media and/or sizes only. Colours, for example, will only berelevantfor discussing paintings, and there are no relevant examples besides those of AjaJ).ta. Equally the six tusks ofthe elephant in Maya's dream can only be depicted if the general size of the scene and the material allows their depiction.7 In the Vinayas ofthe Mahisasaka and Dharmaguptaka schools, which are preserved in Chinese translations only,8 the birth ofthe Buddha is not narrated at alP Instead, these texts focus on the origin of the Sakya-clan, its genealogy and jump directly to the prophecy of the Brahmans. Obviously the Brahmans do not prophecies on the basis ofMaya's dream, since that is not narrated either. In the canonical Pali literature there are only two works that also narrate the birth ofthe Buddha. These are the Mahapadanasutta (MASu)IQ of the Dlgha Nikaya, which is also partly preserved in a Sanskrit version

7 It is likely, that some of the relief sculptures considered here, such as the AmadivatI reliefs, may have been painted originally (Knox 1992: 232). 8 Both Vinayas were only translated in the fIrst half of the fIfth century (Lamotte 1988: 169-170). 9 Bareau 1962: 7-8, 31; Bareau 1995: 16-17, 40. Contrary to the impression left by Lamotte's summaries (Lamotte 1988: 169-170 (1958: 186-187)) these texts do not contain a continuous narrative from birth to enlightenment, but focus on specifIc topics only, namely the origin of the Sakyas and the prediction and events of the Great Departure. IQ Edition: Rhys Davids, Carpenter 1903: 1-54. Translation: Rhys Davids, Rhys Davids 1910: 4-41, Walshe 1996: 199-221. Further: Norman 1983: 36.

43 Christian Luczanits

(Mahavadanasl1tra11 ), and the Acchariyabbhutadhammasutta. Both de­ clare that the events during the last life of any Buddha are, as a matter of fact, identica1. 12 The MASu fIrst cites the individual circumstances of the last lives of Gotama and his six predecessors13 and then only narrates the life - from the descent from Tu~ita heaven to the fIrst sermon - of the fIrst of them, namely Buddha VipassI / Vipasyin. 14 The Acchariyabbhl1ta(dhamma)sut­ ta (ABDSu)15, in the sectiononemptiness (Sufifiatavagga) in the Majjhima­ nikaya, only focuses on the events from getting reborn in Tu~ita heaven to the birth. The parts common to both texts are almost identical. The descriptions ofthese texts contain very little ofillustrative value. A number of incidents are said to invariably occur at the time of the Bodhisattva's descent from Tu~ita into his future mother's womb. The descent is undertaken 'mindful and clearly aware'16, it is accompanied by a radiance illuminating the whole universe, described in great detail, as well as an earthquake,17 and four gods guard the future Buddha's mother in the four directions. 18 Then the Bodhisattva is likened to a gem, his

11 Editions: Waldschmidt 1953, Fukita 2003. Fukita's edition improves the earlier one considerably. 12 Strictly speaking, the Acchariyabbhutadharnrnasutta (ABDSu) does not clarify at all which Buddha's birth it actually describes (Windisch 1908: 103-105). Windisch 1908: 107-137, discusses each sentence of the ABDSu and compares them with other sources on the Buddha's life, namely the Nidiinakatha, Mahavastu and the Lalita.

13 While the events are identical the following circumstances are individual for each Buddha: kalpa, life span, caste, family name, bodhi-tree, number of disciples, principal pair of disciples, attendants, son, parents and capital. The BuddhavaqIsa, focusing on the individual circumstances only, extends this concept to 24+3predecessors of Buddha Gotama (Homer 1975, Jayawickrama 1995).

14 The relevant sections are: Rhys Davids. Carpenter 1903: 11-13; Rhys Davids, Rhys Davids 1910: 8-10; Walshe 1996: 203-204; Fukita 2003: 52--61. 15 Edition: Chalmers 1899: 118-124. Translations: Chalmers 1927: 222-226. Homer 1967-76: rn, 163-169; NiiI)amoli, Bodhi 1995: 979-984. 16 Pali: sato sampajiino. This is not noted in the Sanskrit while it further emphasised by the ABDSu, where also the prior rebirth in Tu~ita and the stay there have been concious.

17 See also Windisch 1908: 110-112. 18 The commentary to the MASu by Buddhaghosa qualifies them as the Four Great Kings (Stede 1971: 434).

44 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art mother's virtues are expounded and the birth is described. There are neither details concerning the sojourn in Tu~ita heaven among the gods nor is the dream of the queen mentioned. 19 Asvagho~a's Buddhacarita20 is a poem that was composed to em­ phasize the moral and intellectual virtues of the Buddha21 and thus does not focus on mythical elements. Consequently, the story commences with praising the virtues of the future Buddha's parents (verses 1,1-2) and the conception. In her dream, Maya sees a white elephant king entering her body (verses 1,3-5). The descent from Tu~ita is not mentioned. 22 Asvagho~a is thought to have lived around 100 CE.23 Somewhat more elaborate on the events prior to birth is the other mahakiivya Asvagho~a' s, the Saundarananda.24 At the beginning of the second canto the virtues ofSuddhodana are extolled (verses 11,1--45). The gods searching an appropriate realm for the Bodhisattva noticed his virtues (verses 11,46-47). The Bodhisattva decides to be reborn there (verses 11,48--49) and, "walking from the Tu~ita gods to earth", takes his

19 See also Windisch 1908: 6,104-105. 20 Edition: Cowell 1893. Edition and translation: Johnston 1936, Schotsmann 1995. Translation: Schmidt 1923. Translation from the Tibetan: Weller 1926-28. Translation from the Chinese version: Beal1883. Analysis: Weller 1939; Weller 1953; de Jong 1990. 21 See Johnston 1936: xxxvii-xl. 22 Nevertheless, there is a sloka of the Buddhacarita narrating the descent from Tu~ita heaven: cyuto 'tha kiiyiit tu~itiit trilokim uddyotayann uttamabodhisattvaf:z 1vivesa tasyiif:z smrta eva kuk~au nandiiguhiiyiim iva niigariijaf:z 11 (see Windisch 1908: 35-36, 166). However, this sloka, 1,19 in Cowell's edition, is part ofthose verses at the beginning of the first canto that do not occur in the Tibetan and Chinese translations (Cowell1893: vi). These verses are thus not considered original and have not not been included in Johnston's more recent edition. The German translation of Richard Schmidt (Schmidt 1923) is based on Cowell's edition and thus contains the apocryphal beginning verses as well (p. 11-13). Interestingly, Weller 1939: 337-38, considers the 'unfinished impression' of the first canto the result of a visual model for the author. 23 See Lienhard 1984: 132 and 64--70 on the author and his poetry. 24 Edition and translation: Johnston 1928; Jamspal1999.

45 Christian Luczanits rebirth in his family.25 In her dream Maya sees a white, six-tusked elephant entering her womb.26 The Nidanakatha (Nd)27, an introduction to the PaIi collection of Birth Stories (JatakaW!avaI).I.lana), narrates the spiritual career of the Buddha in three parts. The Diirenidana (Distant Epoch) covers the previous lives ofthe Buddhabeginning with Sumedhameeting Dlparpkara and ending with the rebirth of Vessantara in Tu~ita heaven. The Avidiire­ nidana (Intermediate Epoch) narrates the actual life of the Buddha until his enlightenment, and the Santikenidana (Recent Epoch) covers the period from the enlightenment to a sudden end after the Buddha received Anathapil.l9ika's gift of the Jetavana. Essentially the text is a streamlined and complemented compilation of sections known from other Pali works, some of them being quoted, among them the Buddhavarp.sa, the MASu and the ABDSu. 28 The compilation of the JatakatthavaJ:.Il.lana probably dates to the fIfth century or around 500, when also the rest of the PaIi canon was fInalized. 29 Concerning the illustrative value, the relevant section may be summarized as follows. 30 The text fIrst distinguishes three types of proclamations, on of them being the proclamation of a Buddha by the guardian deities of the world (lokapiiladevatii). Having heard this procla­ mation, the gods assemble in front of the Bodhisattva and entreat him to become Buddha. The Bodhisattva makes the five reflections (right time, land, place, family, mother) and agrees that it is the right time and dismisses them. With the Tu~ita gods alone, the Bodhisattva enters the

25 devebhyas tu~itebhyo 'tha bodhisattvaJ:t k~iti'!l vrajan 1 upapatti'!l praIJidadhe kuIe tasya mahfpateJ:t 11 Johnston, 1928: II,48. It is apparent that Asvagho~a does not use the usual terminology to describe the descent. 26 svapne 'tha samaye garbham iivisanta'!l dadarsa sii I ~arJdanta'!l viiraIJa'!l ivetam airiivatam ivaujasii 11 Johnston, 1928: II,50. 7:1 Edition: Fausb011 1877: 1-94. Translations: Jayawickrama 1990; Rhys Davids 1998. 28 See Jayawickrama 1990: xi-xv. 29 SeeJ. Dutoit 1921: vol.7: Ill; Nonnan 1983: 77-84,128-30; von Hintiber 1998: 1. There also existed an older version (see, for example, von Hintiber 1998: 41-42). 30 Jayawickrama 1990: 63-67; Rhys Davids 1998: 144-50.

46 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art

Nandana-Park where he passes away and takes rebirth in Maya's womb.3l The fasting queen Maya, lying down on the seventh day, dreams that the Four Great Kings carry her into the Himalayas. There, goddesses bath her and placed her on a heavenly couch in the golden abode of the silver mountain. The Bodhisattva descents from the golden mountain in the form of a white elephant with a lotus in his trunk, circumambulates her trice and enters her. 32 Obviously, the dream is here extended into a side narrative that is essentially out of the actual sequence. Somewhat later it is further stated that from the time of conception the mother is protected by four deities with swords. 33 Continuing by extolling the queen's virtue and the Bodhisattva as gem this section is based on the MASu. TheMahavastu (MV)34,partofthe VinayapitakaoftheLokottaravadin, a branch of the Mahasfu!lghika35, has a structure similar to the Nd. However, the basic narrative ofthe Buddha's spiritual career is interrupted by numerous stories of previous births. The text has older verses in Buddhist Hybrid Sanskrit embedded in and connected by Sanskrit prose. The core ofthe text is said to be very old, but its younger parts, particularly the introduction, are attributed to the 4th century CE.36 First, the MV introduces the (somewhat incomplete) career of a Bodhisattva.37 Then follows the story of the life of Buddha Dlpaq1lcara,

31 so eva1J1 devatiihi kusala1J1 siirayamiiniihi parivuto tattha vicaranto cavitvii Mahiimiiyiiya deviyii kucchismi1J1 pa!isandhi1J1 gal}hi. Fausbl1J1l1877: 50. 32 Windisch 1908: 153-55, Dutoit 1921: 84-90. 33 Jayawickrama 1990: 69. 34 Edition: Senart 1882. Translation: Jones 1949-1956; Leumann, Shiraishi 1957: section on the Bodhisattvacareerand the ten bhiimi; Leumann, Shiraishi 1962: continuing the ten bhiimi section. Analysis: Windisch 1909; Leumann, Shiraishi 1962; Oldenberg 1967; Rahula 1978. 35 Senart 1882: 1,2,13-14, rn, 461,13-14. 36 Wintemitz 1983: 11, 238, postulates that the oldest core may go back to the 2nd century BCE. The final reduction ofthe text is thought to date to the 4th century CE (also Rahula 1978: 16). 37 Thereby two independent systems are connected, namely the four practices of a Bodhisattva (catviiri bodhisattvacaryii':li, these are levels ofthe Bodhisattvadevelopment that are characterised by a certain practice) and the ten bhiimi (Leumann, Shiraishi 1962; Lamotte 1966-80: v, 2373-81, Rahula 1978: 62).

47 Christian Luczanits the fIrst Buddha who prophesied (vyiikaroti) Sakyamuni his future Buddhahood.38 After also Sakyamuni's immediate predecessor, the Buddha Kasyapa, prophesied his Buddhahood he takes rebirth in Tu~ita heaven. Although the deva Svetaketu, as he is called in Tu~ita heaven, excells the other gods, there is no notion that he actually teaches them. 39 The life of Buddha DlpaqIkara is practically identical with that of Gotama, just the names change. Minor differences, especially the shift of a whole group of verses to another place,40 may be the result of the previous oral transmission of the text as is also indicated by many repe­ titions.4l Both texts narrate the same events preceding the birth. Twelve years before his last birth the Suddhiiviisa gods42 proclaim the descent and teach the 32 signs of a great man (mahiipuru~a). Consequently, the Pratyekabuddha leave the earth by rising into the sky and having themselves consumed by fIre. The Bodhisattva makes his four reflections (time, place, continent, family) and decides on a suitable mother. In the meantime the queen has laid herself to sleep and is visited and protected by Rak~asa, Yak~a, and the Four Great Kings ('four lords') and all other kinds ofdeities. The Bodhisattva then departs and during the descent a radiance illuminates the whole Buddha-fIeld (buddhak~etra) and the earth quakes. The gods mentioned above continue to attend to the Bodhisattva and his mother. In both lives, that ofDlpaqIkara and that ofGotama, the actual descent is described in two versions consisting of two verses each:

38 In the career of a Bodhisattva this prophesy marks the threshold between the seventh and the eights bhUmi (Jones 1949-1956: I, 2). 39 "He excelled the other devas in the ten heavenly attributes, namely, heavenly length of life .... This deva named Svetaketu was learned, accomplished, confident, skilled ... " Jones 1949-1956: 1,284.

40 For example Jones 1949-1956: 1,164-67 and 11,9-11.

41 Single verses and groups ofverses from the life have also been used in explaining the tenth bhumi. Among them are even verses that in the actual narration of the life have been replaced by prose (compare, for example, Jones 1949-1956: 1,133 with 11,3 and 1,18 with 11,17).

42 These are the gods in the five highest heavens of the 17 rupadhatu-heavens (see Kloetzli 1983: 29-30, 33-39; Gronbold 1984: 392-395.

48 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

"Lo, I depart. " So did the Exalted One speak out and utter the happy word. Andat that very moment the Conqueror's mother saw in a dream him who had won maturity offruition, Entering herbody in the form ofa noble elephant, light ofstep, flawless of limb, gleaming like snow-white silver, with six tusks, a gracefully waving trunk and a crimson head. 43

And:

When the mighty and mindful one passed away from his abode in Tu~ita, taking on the form ofan elephant ofthe colour ofa snow-white boar, Mindful, self-possessed and virtuous he descended into his mother's womb as she lay abed high up in the palace,fasting and clothed in pure raiment. 44

It may well be that these two versions represent different traditions since also the meter of the verses is different. The second version is the only one where Gotama actually takes on the form of an elephant and descents as such. This 'historicization'45 of the descent is even better visible in another verse that is, without obvious connection, quoted in the description of the third bhumi. There the Bodhisattva descents in the body of an elephant (gajarupavat).46

43 Jones 1949-1956: 11,8 and 1,162. e~o cyaviimi iti mut:tci giraf!! subhaf!! vacanaf!! udfrayi ... 1 atha supinaf!! jananf jinasya tasmif!! k~at:te pasyati varavipiikaphalaf!! 11 himarajatanibho me ~atjvi~iit:to sucarat:taciirubhujo suraktasir~o I udaram upagato gajapradhiino lalitagatitz anavadyagiitrasandhitz 11 Senart 1882: 11,8,15-18 (corresponding to 1,205,1-4). Senart mentions (1,537) that even a comparison of the two occurrences does not allow to close the gap in the fIrst verse. 44 Jones 1949-1956: IT,ll and I, 164. soyaf!! mahiinubhiivo smrtimiif!! tu~itabhavanii cyavitviina 1piit:ttjaravariihakanibho bhavitva gajarilpi ~atjtjanto 11 vfrasayane sayantiye po~adhikiiye visuddhavasaniiye Ism.rta saf!!prajiinakusalo miitutz kuk~ismif!! okriinto 11 Senart 1882: IT,11,18-2l (corresponds to 1,207,7-10).. 45 Windisch uses this term to express the difference of something happening in a dream to an actual event (Windisch 1908: 158,60,66). Itmay well be, that the difficulties to represent a dream visually has ultimately led to the notion that the Bodhisattva actually took on the form of the elephant (Windisch 1908: 173, Foucher 1949: 38, and the depiction of the dream at Bharhut). 46 miiyiiyii devyiitz kuk~ismif!! pravisi~u sa kumudasadrso varo gajarupaviif!! eva ...

49 Christian Luczanits

In the Vinaya of the Miilasarvastivadin (MSV)47 the first part of the life story is told in the Sarp.ghabhedavastu, the 17th and last section. This section begins with the history of the world from its renewal to the genealogy of the Sakya family and the reign of Suddhodana and then narrates the actual life ofthe Buddha. The Abhini~kramaJ:.lasatra,which is only preserved in Tibetan translation conforms with the description in the MSV.48 Following the five reflections (family, place, time, caste and mother) the Bodhisattva informs the gods about his decision and invites them to accompany him. The gods entreat him not to depart, pointing out the erroneous teachings spread all over the world by 18 teachers: 49

"Here, however, [we] prepare a seat for the Bodhisattva to listen to the teaching across twelve yojana. 50 And we think, whatever dharma the Bodhisattva staying in the Tu~ita abode will teach us, that dharma we will practise as we have heard it. In this way [the dharma] will, for a long time, be useful, beneficial and lucky for us. "5/ lokiilokal:t tu~itavarabhavananilaya1Jl vihiiya ihiigatal:t I Senart 1882: 1,98,12-13. Sadly the text is not completely preserved at this point. That this notion also appears here in introductory section of the MV may indicate that it is a fairly late development. 47 Edition: Gnoli 1977. Summary: Rockhilll884. The MSVis the only Vinaya that has been translated into Tibetan, where it is completely preserved. In the relevant sections the Tibetan translations conforms to the Sanskrit version found at Gilgit. See Q 1955-1958: 1030/17, 258a - 261b. 48 This text is not identical with a Chinese text of the same name, which rather relates to the Lalitavistara (T 190, trans. Beal1875). As Beal points out in his introduction, the translation of abhini~krama1}a,referring to the great departure of the Bodhisattva, is frequently used as (part of the) title of Chinese Buddha biographies, including some of the earliest translations. The emphasis expressed by this title on the great departure conforms to the shortest biographies preserved in the Chinese translations of the Vinaya of the Mahlsasaka and Dharmaguptaka schools referred to above. 49 The teachers are mentioned in three groups of six each (Gnoli 1977: 39). 50 In the Tibetan translation (Q 1930/17, 260bl-2) the 12 yojana refer to the Bodhisattva's seat: 'dir na chos mnyan pa 'i slad du byang chub sems dpa'i gdan dpag tshad bcu gnyis pa bsam par bgyi ... 51 iha tu bodhisattvasya dviidasayojaniini dharmasrava1}iirtham iisanaprajfiaptil:t prajfiiipyate; asmiika1Jl caiva1Jl bhavati; yam asmiika1Jl tu~itabhavanastho bodhisattvo dharma1Jl deSayi~yati ta1Jl vaya1Jl dharma1Jl srutvii tathii tathii pratipatsyiimahe; yathiipitad asmiika1Jl bhavi~yati dfrghariitram arthiiya hitiiya sukhiiyeti. Gnoli 1977: 39.

50 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art

The Bodhisattva entreated thus said to the gods of Tu~ita heaven: "Therefore, friends, hit all musical instruments!" After the gods of Tu~ita heaven have hitall musical instruments, andalso the Bodhisattva blew the conch [he said}: "Which of the two, friends, is the highe,s2 sound?" "The sound ofthe conch, bhagavan. "53

"Friends, as this sound ofthe conch is superior to all [other} musical instruments, so am I, after I have descended to earth (jambudvrpa) superiorto the sixdialectics54, the sixreciters55 andthe sixpractitioners56 and will reach deathlessness. I will saturate the world with nectar [= deathlessnessj. I will blow the conch of impermanence. I will hit the drum ofthe void. I will roar the lion's roar ofselflessness. "57

52 odiirika is not to be found in this form in the standard dictionaries. For Edgerton 1953: 161. only the meaning 'gross, material' is secured for audiirika (= Pa1i o!iirika), while he doubts the examples where it may mean 'great'. In Bechert 1994 the latter is given as the first meaning. Regularly derived audiirika should go back to udiira 'high, lofty, exalted' (see PW: 1131). Pali o!iirika "{scdry deriv.from uJiira .. + ika; ..} a.(i) large.. ; extensive ... ; b.(iii) (ofsound) loud, high-pitched" CPD: 767-68..

53 evam ukto bodhisattva~ tu~itakiiyikiin deviin idam avocat: "tena hi miir~ii sarvaviidyiini praha1Jyantiim " iti. tu~itakiiyikiibhir devatiibhi~ sarvaviidyiini prahatiini; bodhisattveniipi smikham iipuryiibhihitam "kataro 'tra miir~ii~ odiirika~ sabda~?" "salikhasabdo bhagavan. " Gnoli 1977: 39. 54 The names ofthe first group correspond to the six 'heretics' in Mhvy 3545-3550 (Tamura, Imanishi, Sato 1966-68). In the MSV this group occurs several times: in the same section when Ajatasatru is converted, in the Pravrajyavastu as part of the story of Sariputra and Maudgalyayana, and in the Vinayavibhanga in the section that discusses the offences that lead to a temporary exclusion from the monastic community. On their teachings see Vogel 1970.

55 This is a group of six Brahmans.

56 Rendered sfioms par jug pa in Tibetan, among them are Udraka Ramaputra and Arac;la Kalama.

57 "yathiiyaf!! miir~ii~ salikhasabda~ sarvaviidyiiny abhibhuyiivasthita~ evam eviihaf!! jambudvfpam avatfrya ~a! tiirkikiin ~alj iinusravikiin ~a! ca pratipattrfn abhibhuyiimrtam adhigami~yiimi; amrtena jagat santarpayi~yiimi; anityatiisalikham iipurayi~yiimi; sunyatiibherff!! tiiljayi~yiimfti; nairiitmyasif!!haniidaf!! nadi~yiimf" Gnoli 1977: 39.

51 Christian Luczanits

Having explained this he spoke a verse: "One lion stirs a multitude ofbeasts ofprey. One vajra cuts many rock peaks. Sakra alone defeats the lords of the Danava. The sun alone disperses darkness repeatedly.

Who among you, friends, desires deathlessness, should take rebirth in the middle country (madhyadeia). G058 to the six large cities [there). "59

Consequently, Sakraprepares the womb ofMahamayaand the Bodhisattva descends in the middle of the night in the appearance of an Elephant.60 Maya has four dreams: a six-tusked elephant enters her,61 she rises into the sky, ascends a high mountain, and a large crowd bows in front of her. The events described here indicate a rather late date for this section of the MSV. The Bodhisattva appears as teacher of the gods. The group of 18 teachers contrasts with the group of the fIrst six that occurs in other places of the Vinaya as well.54 The use of musical instruments as similes and the superior conch, too, may be indicators of a relatively late date.62 The section may, thus, be counted among those that lead Lamotte to the

58 iryati: see Edgerton 1953: 115.

59 iti viditvii giithii'!1 bhii~ate: "si'!1haika~ pramathati naikasviipadaughiin, vajraiko vilikhati naikas[7igasailiin 1 sakraiko vijayati naikadiinavendriin, suryaiko vimathati naikam andhakiiram 11 yo yU~miika'!1 miir~ii amrteniirthr sa madhyadeSe pratisandhi'!1 grhf}iitu; ~a!su mahiinagare~v iryatha. Gnoli 1977: 39-40. The last two sentences are combined in the Tibetan translation (Q 1030117, 161a1). 60 gajanidarsanena riitryii madhyame yiime miihiimiiyiiyii devyii~ kuk~im avakriinta~ Gnoli 1977: 40.

61 ~ar)anto me sveto hastiniiga~ kuk~i'!1 bhittvii pravi~!a~ Gno1i 1977: 40. 62 Music was originally not allowed for monks and the simile may well assume a certain acceptance of music in a Buddhist (monastic) context. However, the usage of music is consistent with the MSV that describes image processions of an image of the Bodhisattva under the Jambu-tree accompanied by musicians. A study ofSchopen 2005: 128-137. indicates that this may well have been the custom during the KU~iiQa period. The emphasis on the conch as the superior instrument may well be connected with the conch that Vi~Qu uses to measure the extend of the ocean.

52 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art conclusion that "we cannot attribute to this work a date earlier than the fourth-fifth centuries ofthe Christian era".63 In an earlier recension the Lalitavistara (LV)64 was a Sarvastivada text, but it has been reworked and extended several times.65 The extant Mahayanasiitra, however, only was finalised in the late seventh or early eighth century and is thus not much older than its Tibetan translation.66 Like the MV it contains verses in Buddhist Hybrid Sanskrit. The LV begins with the sojourn in Tu~ita heaven and ends with the first sermon (dharmacakrapravartana, chapter 26). The description of the Buddha's life is set onto a common teaching frame. In chapter one the Buddha resides with a huge assembly of monks and Bodhisattvas in the Jetavana. At the request of the gods he teaches the Lalitavistara. In the final chapter (27) the Buddha requests Mahakasyapa, Ananda and the Bodhisattva Maitreya to teach this text to others.67

63 Lamotte 1988: 657 (1958: 727). Lamotte notes that in the MSV Sakyamuni is understood as "god superior to the gods" as it is also known from Mahayana texts. Gnoli 1977: XIX-XX., in contrast, attributes the MSV to the period of the Ku~fu?a emperor Kani~ka (20d century). For a more differentiated argument on the date of the MSV see also Schopen 2004: 19-25. I assume here that although, as Schopen has shown, much of the text may well go back to the time of the Ku~aQ.a king Kani~ka the relevant section here is of a later date. In comparison, a teaching Maitreya in Tu~ita heaven is only described in a Chinese source (T 452, translated 455: Lamotte 1988: 706 (1958: 782-783)), but the concept itself certainly can also be proven for the fourth century but, to my knowledge, not earlier. 64 Text editions: Mitra 1877; Lefmann 1902 / 08; Vaidya 1958. Translations: Lefmann 1874 (the fIrst fIve chapters only); Foucaux 1884-92; Bays 1983 (translated from the Foucaux's French translation); Mitra 1998 (fIrst 15 chapters). Analysis: de long 1954; Schubring 1954; Weller 1987; Khosla 1991. Tibetan: Q763. 65 de long studies an example of the revisions and extensions of the text (de long 1954).

66 The suggested date is between the Chinese translation by Divakara (= T 190) 683 CE, which does not yet conform with the extant Sanskrit text, and the Tibetan translation in the early ninth century (de long 1954: 313; Weller 1987: 472 (17)).

67 imiim ahalJ1 miir~a asalJ1khyeyakalpako{iniyutasatasahasrasamudanftam anuttaralJ1 samyaksalJ1bodhilJ1 yU~miikalJ1 haste paridamy anuparidiimi paramaya parindanaya, svayalJ1 caivam imalJ1 dharmaparyayalJ1 dharayata, parebhyas ca vistare1}a salJ1prakasayata" Vaidya 1958: 318,18-20.

53 Christian Luczanits

A Bodhisattva named Svetaketu68 resides in Tu~ita heaven. There resound exhorting verses reminding him of the prophecy of Buddha DlpaIpkara69 (chapter two). In chapter three the Bodhisattva Svetaketu informs the gods about his rebirth in Jambudvlpa. There the gods, disguised as Brahmans, teach the Vedas and announce the rebirth to the Pratyekabuddhas, who leave the earth by rising into the sky and having themselves consumed by fire. The Bodhisattva makes the four reflections (time, continent, place and family) and asks the gods in which family he will be reborn. He explains the 32 characteristics of the mother of a Bodhisattva who takes his last rebirth. Finally he teaches a huge assembly ofBodhisattvas, gods and apsaras for the last time (chapter four). Thereby he teaches the 108 dharmii/okamukha that are taught at the time of a Bodhisattva's departure from the assembly ofgods. 70 Each ofthe chapters is set into a different palace (vimiina). In the fifth chapter, the departure (praca/a), the Bodhisattva proclaims his departure to Jambudvlpa.

Now the gods of Tu~ita heaven, crying, held the feet ofthe Bodhisattva and spoke thus: "Satpur~ua, this Tu~ita mansion will, without your presence, cease to be resplendent. " Then the Bodhisattva spoke to the big assembly ofgods: "This Bodhisattva Maitreya will teach you the dharma." Then, having taken offthe headdress (pa!!amaula, thod dang cod panp from his head, the Bodhisattva placed it on the head of

68 The same name was used in the MV, Ibid., 8,20.

69 smara vipulapufJ.yanicaya smrtimatigatim anantaprajfiiiprabhakarin 1 atulabalavipulavikrama vyiikarafJ.af!! dfpaf!!karasyiipi 11 Ibid., 9,vl.

70 yadavasyaf!!bodhisattvenacyavanakiilasamayedevapar~adisaf!!prakiisayitavyam

11 Ibid., 25,21-22. The dharmiilokamukha conclude with the last three bhumi in the older form (Vaidya 1958: 25,18-21): the avaivartikabhumi characterised by sarvabuddha­ dharmapratipurti; the bhumerbhumisaf!!kriintijfiiina, in which the Bodhisattva reaches sarvajfiajfiiiniibhi~ekatii, and the abhi~ekabhumi, which is characterised by showing all the deeds of a Buddha (10 deeds are enumerated). These stages conform to those mentioned in the MV.

71 As can be concluded from the following ca in the Sanskrit and the Tibetan translation, the headdress consists of two parts: thod dang cod pan (Q 763, 28a8) = *pa.t!o maulaf!! ca.

54 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art

Bodhisattva Maitreya and said: "Without me you, satpuru~a, will attain the highest, complete awakening. "72

Then the Bodhisattva asks the gods in which form he should enter his mother's womb, and after some suggestions a god called Ugratejas says:

"As it occurs in the Mantraveda and Siistra texts ofthe Brahmans, in that appearance the Bodhisattva should descend into his mother's womb. But what [appearanceJ is this? - An Elephant of great dimensions, with six tusks73, covered by a golden net, radiant, with well reddened head and a bursting and dripping body(?Y4. Having heard of such an appearance the Brahman who knows the Vedas and the Siistras will prophesy: 'Indeed, this one will be endowed with the 32 characteristics. '''75

72 atha te tu~itakiiyikii devaputrii rudanto bodhisattvasya cara1}au parigrhyaivam iihuJ:t "ida,!! khalu satpuru~a tu~itabhavana,!! tvayii vihfna'!! na bhriiji~yate" 1 atha bodhisattvas tii,!! mahatf devapar~adam evam iiha "aya,!! [maitreyo bodhisattvo yU~miika,!! dharma,!! de.§ayi~yati" Iatha bodhisattvaJ:tJ svakiic chirasaJ:t pa!{amaula,!! ciivatiirya maitreyasya bodhisattvasya sirasi prati~fhiipayiimiisa 1 eva'!! ciivocat "mamiintarena tva,!! satpuru~a anuttarii'!! samyaksa,!!bodhim abhisa,!!bhotsyase" Vaidya 1958: 28,6-11. In the edition of Rajendra Lal Mitra the section in brackets is replaced by the following: bodhisattvo maitreya,!! bodhisattva,!! yauvariijye prati~fhiipya (Mitra 1877: 43). According Vaidya this edition is based on five manuscripts from India and Nepal (Vaidya 1958: x). Schubring wants to use this gloss and suggests to read the following verb as causative (Schubring 1954: 615).

73 Lefmann 1874: 178-79, suggested that the six tusks refer to the two actual tusks and the four molars that have grown together. See also Janert 1977: 71-72. 74 Since the suggested explanations (for example Schubring 1954: 615) are equally unsatisfying I have chosen a literal translation of this peculiar description (see also Lefmann 1874: 179). Most likely the 'bursting and dripping' refers to an adult male elephant during musth (pers./engl) when a secretion originating at the temples runs down the sides of the head (see Janert 1977: 69-70).

75 yathii briihma1}iinii,!! mantravedasastrapiifhe~v iigacchati tii drsenaiva rupena bodhisattvo miituJ:t kuk~iiv avakriimitavyaJ:t ItatpunaJ:t kfdrsam? gajavaramahiipramii1}aJ:t ~a#anto hemajiilasa,!!kiisaJ:t suruciraJ:t suraktaSfr~aJ:t sphufitagalitarupaviin I etac chrutvii rupa,!! briihma1}avedasiistratattvajfio vyiikar~itas ca 1 ita vai bhiivf dViitri,!!sa!!ak~a1}opetaJ:t11 Vaidya 1958: 28,16-20.

55 Christian Luczanits

Finally, the Bodhisattva is venerated by countless gods and departs from Tu~ita heaven. In the presence of the gods he takes place on the lion throne SrIgarbha.76 Countless gods carry the great palace and apsaras praise the Bodhisattva in their singing.77 Chapter six, the descent into the womb (garbhiivakranti), sets the descent on the 15th day of full moon. The Bodhisattva becomes a young elephant with six tusks, a reddened78 head and golden teeth and enters his mother from the right side.79 Maya sees the elephant in her dream80 and describes it in a verse. 81 I have not dealt with the numerous non-canonical Chinese sources on the life of the Buddha since they appear to provide no information that is relevant to the argumentation ofthis article. Unless mentioned otherwise, these sources have only been available to me in the form of text analysis provided by Dieter Schlingloff and secondary sources.82 However, these are not detailed enough that they can be used to develop a more com­ prehensive account of the development of the textual sources. All texts appear to contain Tu~ita heaven, at least in the form of the reflections, and the dream of Maya. 83 It would be interesting to know,

76 See Vaidya 1958: 38,30-39,1. 77 See Vaidya 1958: 39,24-26.

78 The term indragopaka denotes an insect used in making a red dye. 79 iti hi bhik~ava: sisirakiilavinirgato vaisiikhamiise visiikhiinak~atriinugate rtupravare vasantakiilasamaye taruvarapatriikfT1}e varapravarapu~pasa,!!kusumite sfto~1}atamorajo vigate mrdusiidvale susa'!!sthite tribhuvanajye~!ho lokamahito vyavalokya rtukiilasamaye paficadasyii,!! pun.wmiisyii,!! po~adhagrhitiiyii miituJ:z pu~yanak~atrayogena bodhisattvas tu~itavarabhavaniic cyutvii smrtaJ:z sa,!!prajiinan pii1}ljuro gajapoto bhutvii ~aljdanta indragopakasiriiJ:z suvaT1}ariijfdantaJ:z sarviifigo 'hfnendriyo jananyii dak~i1}iiyii,!! kuk~iiv avakriimat Vaidya 1958: 43,1-6.

80 himarajatanibhas ca ~aljvi~ii1}aJ:z sucara1}a ciirubhujaJ:z suraktasfr~aJ:z 1 udaram upagato gajapradhiino lalitagatir drljhavajragiitrasaf!ldhiJ:z 11 Vaidya 1958: 43,8-11. 81 himarajatanikiisas candrasuryiitirekaJ:z sucara1}a suvibhaktaJ:z ~aljvi~ii1}o

mahatmii 1 gajavaru drljhasa,!!dhir vajrakalpaJ:z surupaJ:z udari mama pravi~!as tasya hetu'!! s!1}u~va 11 Vaidya 1958: 44,12-15 (v 6), same as 45,3--6 (v 10). 82 For example the summaries and references found in Beal1883: xvi-xxx; Bareau 1962: 202-205, or Bareau 1995: 46-49). 83 Possibly, this is also valid for the earliest text on the Buddha's life translated 68-70 CE (BealI983: xvi-xvii). The text itself, however, is not preserved.

56 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

when the Bodhisattva fIrst occurs as the teacher ofgods in Tu~ita heaven.84 Itis remarkable, thatnone ofthe Chinese sources mentions the transference ofthe crown to Maitreya as it is narrated in the LV. 85 In China the descent itselfis depicted showing the Bodhisattva riding the elephant. Apparently, the Chinese is ambiguous at this point.86 Although issues ofthe chronology and the relationship ofthe sources are certainly problematic, the Indian textual sources clearly indicate a development for the events prior to birth. As Bareau has worked out in great detail, the canonical sources themselves undergo a considerable development in the fIrst centuries after the Buddha, the birth and the events preceding it being not mentioned at all originally.87 The canonical sources that do mention the descent are interestingly those that do focus on the group of seven historical Buddhas and not on Sakyamuni alone. These texts establish that the events of any Buddha's life follow a set pattern. Further, it is not the descent itselfthat is described but the miracles that occur with it and indicate the conception of a very special being. In what is probably the earlier strata of an actual event taking place in Tu~ita heaven, the Bodhisattva is requested by the gods to be reborn (Nd) or decides on the right time himself (MV). The gods accompany and protect him, respectively his future mother, during the descent. In a further development, the Bodhisattva teaches the gods in Tu~ita heaven (MSV, LV). Consequently, the role of the god changes from supporting the rebirth to trying in vain to prevent it. In the MSV they point out the heretics on earth and that teaching them in heaven would be much more effective, in the LV they emphasise the loss for the heavenly abode. The Bodhisattvareacts quite differently at this stage, in the MSV he encourages

84 I once noted down that Taisho 185, a Buddha-biography translated between 222 and 280 CE describes the BS such, but am unable to control this. If this is indeed the case, this would be surprisingly early. Maitreya appears not to have been understood as reaching there before the fourth century. 85 Foucher 1905-51: I, 232, n.2; Foucher 1949: 36. 86 See Beal1983: xix, n.!. 87 Bareau 1962, and Bareau 1974. Accordingly, not even the personal name of the Buddha's mother is known, since Maya derives from mata, the term for mother (see von Hiniiber 1994: 13-14).

57 Christian Luczanits the gods to become reborn with him while in the LV he appoints a successor to replace him in heaven, namely Maitreya. In the more embellished stories, the description of descent oscillates between the Bodhisattva's active role in it and the dream of his future mother, the latter signifying both the descent and the conception. While in the earlier stratum oftexts it is clear that the descent and Maya's dream are the same event viewed from different angles, in later sources a sequence of events develops that increasingly separated the departure from Tu~ita heaven and the dream. Only one set of verses in the MV explicitly lets the Bodhisattva take on the form of an elephant, while the LV lets him become an elephant in the narration of the dream sequence only. Such ambiguity in the textual sources can in part be referred back to the oral tradition these accounts are based on. In fact, much ofthe relevant information is found in the verse sections of the text. Although in the texts these verses, too, have been edited and reformulated,88 they are generally thought to represent an older stratum of the text. It is quite likely, that these verses have been handed down orally. If one can speak of a textual (including oral) tradition concerning the events under discus­ sion at all, it is the verses that represent it. Although the events prior to birth become increasingly elaborated, relatively little ofthe narration has illustrative value. In the most elaborate form the sequence of events may be summarised as follows: the Bodhisattva stays in Tu~ita heaven (Nd, MV, MSV, LV), the gods announce his impending rebirth on earth and the Pratyekabuddhas immolate themselves (MV, LV), the Bodhisattva communicates with the gods and/or teaches them (MV, MSV, LV), he crowns his successor Maitreya there (LV), he descends (MASu, ABSu, SN, Nd, MV, LV), Maya dreams ofan elephant entering her womb (BC, SN, Nd, MV, MSV, LV) and descent and/or queen are protected by gods (MASu, ABSu, MV, LV). ThedescriptionoftheelephantinMaya'sdreamcontains considerable illustrative details, in one account he is holding a lotus in his trunk (Nd). Only in the LV is the descending Bodhisattva transported in a palace.

88 See, for example, the study of Windisch 1909.

58 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

The succession of the sources documents a development that can be paralleled with the increasing incorporation ideas associated with Mahayana Buddhism, but does not strictly represent a chronological succession. In fact, the problems concerning the chronology of the texts and the temporal, geographic and doctrinal ranges they were used in are too complex to be treated appropriately here.

Visual Sources While in the case of a text, its content can essentially not be contested once the language and the vocabulary are established, the appropriate evaluation of visual sources is much more problematic, since much of it depends on interpretation. This is even more so the case, when the episode is rather secondary and not enclosed in a clear narrative sequence. The Buddhist narratives discussed in the works of Alfred Foucher, especially those on Gandharan narrative art, are exemplary and still a point of reference.89 In the absence of captions his interpretations are based on comparison to the literature and the depiction of similar compositions and motives. Goal has been to identify as many of the depicted narratives as possible. His comparative approach has also been decisive for NJ. Krom's monumental analysis of the narrative reliefs of BarabU<;lur9O, Dieter Schlingloff's extensive studies of the narratives in the art ofAjal)ta, including those ofthe Life ofthe Buddha,91 and Deborah E. Klimburg-Salter article on Tu~ita depictions.92 Focusing on the modes of visual narration Vidya Dehejia, too, follows the traditional identifi­ cations and attributions in all cases discussed here.93 Concerning the present topic Schlingloff notes: "The episode is recounted in all the texts without any variations ofcontent that are relevant

89 Foucher 1905-51: 1,285-89. Due to his familiarity with the textual sources, he even considers episodes, such as the giving of the crown to Maitreya, that have not been depicted in the narrative depictions he studied. 90 Krom 1925; Krom 1927. 91 Schling10ff 1988: 1-48; Schling10ff 2000: 39-71,294-380. Most remarkable are the extensive bibliographic references to each scene offered by these works. 92 Klimburg-Salter 1988. 93 Dehejia 1997.

59 Christian Luczanits

for the pictorial representation. ''94 It is my hope that my analysis of the textual sources above and the visuals following demonstrates that this is not the case once one abandons a purely comparative viewpoint in the favor of a more contextualised approach. The earliest relevant depictions are preserved from the Bharhut railing. On the basis of his interpretation of the inscription, Heinrich Lliders identified the middle scene on the west gate, the Ajatasatru / Ajatasattu pillar (left side)95 as the gods informing the Bodhisattva in Tu~ita heaven that it is time to become reborn.96 The inscription of this panel says: Mahiisiimiiyikiiya Arahaguto devaputo vokato Bhagavato siisani pa!isa1J'ldhi and is translated by Lliders: "Descendedfrom (the hall of) the GreatAssembly, the angel [for devaputo] Arahaguta (Arhadgupta) appraises the Holy One of(his future) incarnation."97 I am certainly not in the position to discuss the inscription as such, but the way Lliders bases his translation on the depiction itself certainly indicates that its interpretation is problematic.98 Neither is a heaven mentioned in the inscription nor has the name of the god, a certain Arahaguto, been found anywhere else but in Bharhut.99 There, the same god accompanies the Bodhisattva at the great departure. 100 Currently I can not offer a conclusive alternative interpretation of the scene, but a consideration of its context may help to place the event. The scene above Arahaguto's visit to the Buddha shows the turban venerated in TrayastriJ11sa heaven in the course of a cutjii feast. 101 The scene below,

94 Schling10ff 1988: 16. The interpretation ofthe relevant episodes has not changed in his more recent study (Schlingloff 2000: 39-51). 95 Cunningham 1879: pI. xvi. 96 An alternative identification, albeit not convincing either, has been suggested by Barua 1979: 14-16.

97 Ltiders 1941: 170-172; Ltiders 1963: 87-88. This interpretation is also followed in Sch1ingloff 1988: 4, Sch1ing1off2000: 46, and Dehejia 1997: 88-89.

98 "The interpretation ofthe inscription is to be based on the sculpture." Ltiders 1963: 87. 99 See also Dehejia 1997: 106.

100 Cunningham 1879: pI. xx. Dehejia 1997: 92-94, fig. 66.

101 See MV, lones 1949-1956: 161. At Bhiirhut, the scene is identified by two inscriptions.

60 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art identified as king Ajatasatru's veneration of the Buddha, gave the pillar its name. 102 The other face of this entrance corner pillar of the West gate shows the miracle of SravastI and the descent from the Trayastri:rp.sa heaven, equally in three scenes. 103 Thus, on both narrative faces of the pillar the Trayastri:rp.sa heaven takes the upper position, while the lower position is occupied by an earthly event. Indeed, if the scenes at Bharhut are related to each other, they are always arranged in a spatial hierarchy. 104 As such, it is unlikely that a Tu~ita scene is placed on a middle panel and that the hall or palace supposed to be mentioned in the inscription is not depicted. Further, there are also two winged deities among the gods in the scene that contradict a Tu~ita identification. If the scenes depicted are somehow related, then the event took place after the great departure, since the headdress is already in heaven. If we exclude the identification of this event, the earliest examples of Buddhist narrative art only feature the dream of Maya prominently. The famous depiction of the dream in a medallion of the Bharhut railing is even identified by inscription: bhagavato ukra'!lti - "the descent ofthe Bhagavat ".105 In this famous depiction a huge decorated elephant hovers above the sleeping queen lying on a bed. 106 The decoration ofthe elephant, a headband with tangles and bells at the side of the head, signifies that this is a royal elephant rather than a wild animal. 107 The queen is attended by three maids, the one in the background standing in veneration, the other two seated in front of the bed, with the back towards the viewer,

102 Contrasting to Liiders' claim this pillar alone is sufficient proof that heaven and earth can not be differentiated on the basis ofthe composition or the symbols the Buddha is represented by (see Liiders 1941: 171).

103 Cunningham 1879: pI. xvii. Dehejia 1997: 13-14, fig. 8 shows both faces of the pillar. 104 Also the Sancl narratives are dominated by their spatial arrangement. 105 Liiders points out that the ii- cannot stand for the usual ava- but may replace upa-, in this case the bhagavat would be approaching rather than descending (Liiders 1941: 45; Liiders 1963: 89).

106 Cunningham 1879: pI. XXVIII. Harle 1994: fig.6. Sharma 1994: 23. Dehejia 1997: fig.57. 107 At Bharhut this distinction is clearly made (compare, for example, Cunningham 1879: pls.xii, xv, xix, xxv, xxvi; Dehejia 1997: figs. 1, 32, 57, 62, 82).

61 Christian Luczanits attending with a fly whisk and possibly an offering. 108 Curiously, both sitting attendants have their left hand raised to the ear. A lamp placed at the feet of the queen indicates that this is a night scene, and a pitcher stands at the floor to the side of the queen's head. Except for the fragmentary depiction of the great departure, no other scenes of the Buddha's youth can be regarded as securely identified among the Bharhut narrative relief sculptures. 109 A similarly huge elephant is featured in the depiction of the dream at SiificI as well. There, the scene is found on the upper edge of the large central panel on the inner face of the eastern gateways north pillar. It depicts the return of the Buddha to the city of Kapilavastu with a royal procession moving towards a place where the Buddha trod (cankama)11O outside the city. On some reproductions, two maids can be recognised to the sides of Maya's bed. III The panel immediately above shows one of

108 I feel that the gesture of this attendant is too ambiguous to conclude that it is an expression of astonishment at the sight of the elephant (this has been suggested by Ltiders 1941: 45-50, and is followed by Dehejia 1997: 86) and this indicates that here an actual event is depicted. 109 I also do not follow Ltiders' identification of the dancing apsaras on the south gate pillar, the so-called Prasenajit pillar (Cunningham 1879: pl.xv. Dehejia 1997: fig.34), as standing for the birth of the Buddha (Ltiders 1941: 52-62). The traversing pillar shows that the three panels on the pillar have to be read together and thus refers in its totality to the Buddha Sakyamuni's enlightenment. The Buddha's enlightenment is represented by his enshrined bodhi-tree, underneath this, the gods of the different heavens, including an emphasised demon and a frustrated Mar-a, refer to the story of the enlightenment. The dancing apsaras below celebrate the event with dance and music. As such it represents a curious mixture of narrating an event ofthe past and documenting the actual place and customs of worship at a later time as it is characteristic for early Buddhist narrative art. Dehejia 1997: 44--46, reads the story in the same way, but emphas­ ises the past event of enlightenment over the veneration of the place at later times. However, the visit of the gods, the frustration of Mara as well as the performance of the apsaras can also be interpreted as referring also to the veneration depicted in the uppermost panel. 110 See Ltiders 1941: 32-40.

III Marshall, Foucher 1940: vol.2, pl.50a. In the text the authors state (voU, p. 202): "it now serves as a notice that the principal place ofthe scene is Kapilavastu." Further: Fergusson 1873: pl.xxxiii; Sivaravamurti 1974: fig.413. Rao 1994: P.lO.

62 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art those almost generic veneration scenes with gods assembled around a seat in front of a garlanded tree. 112 No scene ofSane! has been identified as representing the Bodhisattva in Tu~ita heaven, the same pillar that shows the conception also features the successive heavens, each of them with identical features. Each of the heavens is actually a palace with the gods seated between its columns. The central god sits in lalitiisana and holds a thunderbolt (vajra) and a flask. 113 Although reliefs dedicated to the life of the Buddha are frequent in Gandharan art, there is only one that presumably shows the Bodhisattva in Tu~ita heaven, namely a unique panel of the Sikri Stiipa (Lahore Museum). 114 In this relief, the main image wears a turban, a necklace and earrings together with monastic dress, and he is seated in meditation on a lotus with down-turned petals. Four turbaned and bejewelled figures flank the central image in the foreground, four more with hair-knot and nimbus are in the background. All of them direct the veneration gesture (afijalimudrii) towards the central image. I 15 There are a number of other depictions in Gandharan art that have been identified as representing the Bodhisattva in Tu~ita. However, due to the variety of the depictions and the complexity of the questions involved they need to be studied separately and in greater detail. II6 Here, only the main argument can be summarized.

112 On the basis of the tree, a Nyagrodha lficus indica orficus benghalensis /inn), the scene is identified as depicting the instigation of the gods to teach (adhye~a1Ja; see Marshall, Foucher 1940: vol.!, 210 and vol.2, pl.50a).

113 Marshall, Foucher 1940: pl.49, a+b. Dehejia 1997: fig.97.

114 This identification has been suggested by Foucher 1903: 248-253, and no better explanation for the relief has been suggested since. On Sikri see also Alam 1994; Dar 1999/2000: 19--43. 115 Ingholt 1957: fig.8; Marshall1960: 79-80, fig.74; Dehejia 1997: fig. 161. The style of the Sikri reliefs is highly unusual. The emphasis of the central figure in size and posture prefigures the later iconic representations.

116 I have already dealt with the relevant Gandhiiran depictions that are found within a narrative context (Luczanits 2005). The wider issue of Tu~ita heaven representations and their relationship to Bodhisattva Maitreya is the subject ofa separate monograph I am currently working on.

63 Christian Luczanits

Often depictions of a Bodhisattva holding a flask in his left hand are interpreted as Tu~ita scenes. Most frequently, the Bodhisattva is identified with Maitreya, but some authors also consider him as Sakyamuni. 117 The relevant depictions show the central Bodhisattva seated in meditation, displaying abhayamudrii with the left hand and holding the flask (kalasa, kama1}ljalu) in the right. He has his hair bound in a knot, wears jewellery and often has a nimbus. He may be seated on a simple throne covered with a textile, a footstool in front of it, and underneath an umbrella or baldachin. Alternatively his throne is a lion-throne (si7Jlhiisana) with a baldachincarriedby pillars to whichcherubs with fly-whisks are clinching. Further, the Bodhisattva is surrounded by a number ofpersonages, mostly wearing a turban, of which at least one makes a gesture of conversation or discussion. There are two examples of such a scene within a fairly clear narrative sequence. Well known is the panel from the Musee Guimet, Paris, attributed to southern Swat. 118 This relief features two scenes, to the right the birth of the Buddha, to the left the discussion scene in a paradisal setting indicated by the balconies to the sides of the central pavilion. Probably less well known is the false window gable from the Cleveland Museum ofArt that contains five scenes to be read from top to bottom. 119 These are, the birth, the discussion scene, the Bodhisattva amusing himself in the harem, the Bodhisattva leaving the sleeping women and the great departure. 120

117 In particular Lobo 1987; Lobo 1991.

118 MG 26360 (fonnerly in the Louvre), I: 58,5 cm, h: 22 cm, d: 9,1 cm. Published in Foucher 1905-51: I, fig. 164; Lobo 1987: abb.8; Kurita 1990: 24, fig.45; Lobo 1991: no.2, fig.2; Gies, Cohen 1996: 233-234, no.I77.

119 Inv. no. 58.474, gift of George P. Bickford, h: 68 cm, w: 41,3 cm. Published in Marcus 1967: 254, fig. 13; Czuma 1985: 188-89, no.0l. 120 Other relevant reliefs are, for example: - Munich, Volkerkundemuseum, no. 12-57-1 (Sarnmlung Cassirer), "Bituminoser Tonschiefer", h: 26cm; published in Schennan 1928: abb.17; Meunie 1942: pl.XIII, 46; Lommel 1974: fig.3; Mallebrein 1984: 29-30, no.15 (identified as Maitreya); Lobo 1991: no.l, fig.1. - Taxila, Dharmarajika Stiipa, Taxila Museum, phyllit, h: 8 3/4in; published in Marshall 1916: 16, pl.IX, e; Meunie 1942: pI. XIII, 47; Ingholt 1957: fig.37; Marshal11960: 102, Lobo 1991: no.6, fig.5.

64 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

While the succession of the depicted events can - and has been ­ disputed for the fIrst example, there is no doubt from the second one that the relevant scene is chronologically to be placed between the birth and the great departure. I have discussed these and other relevant depictions at length elsewhere115 and came to the conclusion that the only way these scenes can be read without making Sakyamuni and Maitreya icono­ graphically indistinguishable is to interpret them as depicting the life of Maitreya, rather then Sakyamuni. The relevant scene would then have no narrative function as such but would serve as the indicator that here the life ofMaitreya is depicted. Thus, the scene refers to the paradisal context of KetumatI, the city in which Maitreya is to be reborn. To come to the above conclusion I also studied how the life of the Buddha actually begins in Gandharan art. Not unexpectedly, possible Tu~ita scenes are much more rare than depictions of Maya's dream. l2l Further, Maya's dream is often shown in combination with the prediction of Buddha Dlparpkara. Of those friezes that have scene(s) between the dream and Dlparpkara, only one can be interpreted as Tu~ita. 122 Thus, the life of the Buddha in Gandhara commonly begins with the story of Dlparpkara, rarely another Jataka follows, 123 and the last life then commences with Maya's dream. 124 Rather than plainly focusing on the

121 In Butkara we find the dream of Maya occurring five times, the interpretation of the dream four times, the birth ten times and only two panels with a flask-bearing Bodhisattva venerated by attendants (WS.60, WS.107, Faccenna 1964). The two latter panels are from the collection ofthe late Wall Saheb ofSwat (information received from Maurizio Taddei). 122 Japanese private collection, Kurita 1990: I, no. 17. In this scene a moustachioed figure, seated on a throne with the knees wide apart and heels touching each other, holds a bunch of flowers in his right hand. Two venerating male figures sitting in the same pose flank the main image. Schlingloff(Schlingloff 1988: 304-307) suggests identifying the scene as representing the Bodhisattva in Tu~ita heaven. Indeed, if this is a Tu~ita scene, the unusual iconography of the main figure shows that the depiction ofthis topic was not well established.

123 Dar 1999 I 2000 notes that besides the DIpaqIkara depictions no Jatakas are found in Sikri.

124 Patricia Eichenbaum Karetzky 1992 begins her account of the Life of the Buddha with DIpaqIkara followed by the nativity.

65 Christian Luczanits story of the Buddha's last life these depictions emphasise the spiritual development of the Buddha as Taddei repeatedly pointed OUt. 125 Independent of the narrative of the life of the Buddha, Gandharan art also shows the Bodhisattva Maitreya in Tu~ita heaven. These depictions may even be iconographically identical to the discussion scene but also somewhat simpler. It is not clear when the Tu~ita representations came up, but it can be considered quite certain, that by the end of the second century, Maitreya's presence there was established. The relief sculptures of AmaravatI and Nagarjunakol).<;la are consid­ erably more elaborate than those prior to the Common Era. 126 By the time the major Andhra reliefs were carved antropomorphic Buddhas were already depicted in the Northwest. 127 Dating from the second to the fourth century, the Andhra reliefs oscillate between the symbolic representation of the Buddha, as was the rule for the earliest Buddhist art, and the depiction ofhis image. This alone allows to expect that the conception of the Buddha's life has considerably changed since the earliest depictions. In fact, the A.ndhra reliefs do not only depict the Buddha-to-be in Tu~ita heaven, but there is also a separate scene for the descent independent of Maya's dream. Besides the short birth cycle beginning directly with Maya's dream128 or the birth l29, there are occasional depictions of an extended cycle in the later reliefs ofAmaravatI and Nagarjunakol).<;la. I am focusing here on two exceptional renderings of the topic in the AmaravatI and use those of Nagarjunakol).<;la only in comparison, the fIrst one to introduces the ex­ tended cycle and the second one to show the innovative nature of the narrative reliefs at the site. Most of the relief sculptures from the Great Stupa at ArnaravatI are attributed to the second and third centuries on the

125 Taddei 1993; Taddei 1995; Taddei 1999.

126 No scenes of relevance have been found at the smaller sites of the region.

127 In the Northwest the transition from the symbolic representation of the Buddha to that of his image took already place in the 1 century CE. 128 For example: Amaravatl: Rosen Stone 1994: fig.84, fig.189 (four scenes on a square drum slab), the latter being the same as Knox 1992: no.61.; NagarjunakoQ.<;la: Rosen Stone 1994: fig.36 (details 41 and 42, site 6), fig.65 (details 83, site 9).

129 The ayaka-platform friezes Rosen Stone 1994: fig.158, 209, 28.

66 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art basis of inscriptions mentioning the Satavahana kings, the first of them being Sri Vasi~thlputra Pu!umavi (first half ofthe second century).130 The Satavahana are followed by the Ik~vaku dynasty based at Nagarjunakol).<;la and ruling from c. 225 beyond 300 CE. The fust example which I will discuss is a coping stone now in the Indian Museum, Calcutta, with three scenes to be read from left to right, the narration thus being represented on the inner face of the railing. 131 Line drawings ofthe three scenes have already been utilized by Fergusson, Burgess, Foucaux and Foucher. 132 Thefustsceneillustrates the Bodhisattva seated on his throne in Tu~ita heaven. His right hand is raised performing an argumentation gesture while the left rests on the thigh. He wears the same type ofjewellery that is also used for the other gods. The 17 gods surrounding him mostly make gestures of veneration, just the two gods flanking the thrown brandish a fly-whisk (ciimara). The gods in the foreground are seated, while those in the background stand. Between those kneeling in front of the throne is a bowl of flowers(?).133 The succeeding scene then shows the actual descent. A host of gaIJa carries a palanquin with the Buddha-to-be in the form ofan elephant. 134 The accom­ panying gods and gaIJa play music and dance and one of them carries an

130 Knox 1992: 13-14.

131 Sivaravamurti 1974: fig.416; Dehejia 1997: fig. 140.

132 In one drawing: Fergusson 1873: pl.lxxiv. In the drawings the relief is shown in three parts: Foucaux 1884-92: 42, 50, 55; Burgess 1886: figs.7, 17,65; Foucher 1905­ 51: figs. 146-48. 133 In a NagarjunakoI).~a example the Bodhisattva performs abhayamudra and has a nimbus (Longhurst 1938: pl.XIX, c ("The exhortation of the gods to the Bodhisattva") and d; Snellgrove 1978: fig.50; Klimburg-Salter 1988: fig.6; Rosen Stone 1994: fig. 192 (attributes the panel to site 3)). The descent alone is depicted in Rao 1956: pl.XLIV, called "Mayas Dream"; and Parimoo 1982: fig.2A. The continuation of this relief is in the National Museum, New Delhi, and shows the interpretation of the dream, birth, visit to temple and Asita (Sivaravamurti 1974: 74). Sivaramamurti 1942: pl.XXX, 1, shows another example for the descent at Amaravati. 134 A miniature palanquin may also be represented in Maya's dream of NagarjunakoI).~a site 6 (Rosen Stone 1994: detail fig.42, site 6), but it is definitely recognizable in the two ayaka-frieze representations of the newly excavated and unpublished site of Kaganhalli, Gulbarga (photos WHAV CLOO 39,05 and CLOO 42,09­ 11).

67 Christian Luczanits umbrella. Thethird scene then depicts the dream ofMaya, herbed guarded by four gods or the Four Great Kings. 135

The continuation of this story is preserved in the British Museum. 136 Again divided into three scenes it shows the interpretation of the dream, the birth and a scene in the palace. 137 In the birth scene the Bodhisattva is shown symbolically only, seven footprints coverthe cloth and an umbrella hovers above the fIrst ofthem. While the Bodhisattva is represented in his heavenly and miraculous forms, he is not shown as a human. These reliefs are attributed to the later phase of AmaravatI and thus to the third century. 138 The second, highly exceptional relief from the Great Stiipa at AmaravatI, is dedicated to the descent alone. 139 Here, the descent is spread over four separate fIelds on the top of a railing pillar. The main scene in the central flute underneath the top half lotus shows the elephant being carried to earth in his palanquin by a group ofga1}a and attended by gods. The half lotus above represents Tu~ita heaven. In the centre is the now empty throne with the footprints of the Bodhisattva on it and a tree to its back. Three gods each, mostly with gestures of veneration, sit to the side ofit. Underneath them four more gods, also with aiijalimudrii, are looking after the descended Bodhisattva or have decided to accompany him and dive towards the central scene. The fluted panels at the sides underneath the lotus contain more of the attending gods, some diving towards the

135 Amaravati: Burgess 1886: pI.xxviii,2; Sivaravamurti 1974: 97. NagfujunakOJ;I<;la: Longhurst 1935: 149-50, pI.XXXIX, (t) (horoscope and birth following); Fabri 1936: pI.XLV, (d), fragment, description: "Probably the mahiibhini~krama1Ja of Prince Siddhartha". 136 Knox 1992: 100--101, no.41, Dehejia 1997: fig.141. The way the scenes are divided and the decorative bands on the upper and lower edge proof that these two reliefs succeed each other. 137 According to Knox 1992: 101. this scene represents prince Siddhartha with his harem. This would mean that there are no further nativity scenes. However, this may rather be the adoration ofthe infant Buddha as Dehejia maintains (Dehejia 1997: 162). 138 Knox 1992: 101.

139 Burgess 1886: pI.xi,l, Sivaramamurti, "Amaravati Sculptures in the Madras Government Museum," pI. xxx, 1; Burgess 2005: p1.13 (slab no.139).

68 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art others, some oriented towards the pavilion as adoring spectators and some moving along with the pavilion towards earth. In AmaravatI such an extended birth cycle is further shown in the small medallions decorating the dome of the Stiipa. For example, Stiipa reliefOA 1880.7 - 9.87 shows four scenes preceding the birth, the dream of Maya in the centre being 10st. l40 Independent of the birth cycle de­ pictions, Knox also identified a number of reliefs on the front face of the ayaka-platforms as representing the Bodhisattva in Tu~ita heaven. 141 If this identification is right, the scene itself is a very generic veneration scene, its emphasis on the front face of the ayaka-platform would be surprising. There is actually no ayaka-frieze yet known that shows the Tu~ita heaven. 142 Concerning the Tu~ita scene, it is unclear if the mudra performed by the Bodhisattva in the Amaravati example is to be understood as vitarkamudra in the sense of a teaching gesture. In the comparative examples the Bodhisattva performs the more usual abhayamudra. 132 The notion that a Bodhisattva has the ability to teach the dharma presumes rather developed forms of ideas associated with Mahayana Buddhism which are not yet evident in the MV. Thus, Foucher'sl43 interpretation of the Amaravati scene as representing the teaching Bodhisattva in Tu~ita heaven may be far fetched. More likely it is to be understood as a general gesture of discussion, such as the Bodhisattva informing the gods about his decision to get reborn and the result of his investigations.

140 Knox 1992: 149-151, no.77. This relief is unusual in that the central scene was the dream of Maya, in other examples it is commonly the birth that is depicted in the medallion above the entrance (Knox 1992: no.68(?), 70, 71, 72, 75). Knox 1992: no.74, in contrast, may show the same events as no.77. 141 Of these the identification of the respective scene in Knox 1992: no.69., and 78 appear plausible. For the roundel of stupa slab no.76 the identification is impossible since the Bodhisattva is shown meditating. Slab no.75 the central scenes are identical to those of 76, with the exception that the Bodhisattva shows abhayamudrii. 142 Knox 1992: no.37, also identified an iiyaka-frieze scene as Tu~ita, but the continuing narrative and the occurrence of women in the scene exclude this identification. 143 Foucher 1905-51: 288-289.

69 Christian Luczanits

In AjaI.l!a the Tu~ita episode is found in two painted cycles of the Buddha's life, those of caves 2 and 16,144 both attributed to the second half of the fifth century.145 The two depictions are quite similar, but only the one in cave 2 is preserved in its entirety. 146 Cave 2 preserves only the birth events up to the presentation of the newly born to a yak~a, while cave 16 contains the story up to the events just before enlightenment. 147 According Schlingloffboth cycles follow the description in the MSV. 148 In cave 2 the scene is placed in an open square pavilion that also separates it from others. The Bodhisattva sits frontally on a throne with backrest, both legs pending from the seat in European fashion (bhadriisana). He occupies approximately a third of the available space. His hands are in dharmacakramudrii, he wears a knee-long dhoti and the jewellery of a Bodhisattva. His high crown distinguishes him from the other gods. He has a nimbus and the cushion of the throne back forms an aureole. The sides of the backrest are decorated with makara-heads. The Bodhisattva is flanked by four bejewelled gods sitting with their legs crossed at the ankles on cushions. The two in front perform aiijalimudrii and the ones in the back gestures of communication. There are two more attendants with fly-whisks. The parallel depiction in cave 16 has the same elements, but the scene is less separated from he other ones. The gods sit on flat mats or directly on the floor. In both cycles the Tu~ita depiction is directly followed by the dream of Maya.

144 I do not count the life of the Buddha in cave 10 among these. Schlingloff assumes that the life of the Buddha there started with a depiction of the Bodhisattva in Tu~ita heaven (ScWingloff 1988: 4; ScWingloff 1999: pls.23-24; Schlingloff 2000: 39-46). However, the very few remains of that scene speak as much against a Tu~ita identification as against a dream ofMaya. Neither the woman attending with a fly-whisk nor the man with a helmet-like headdress are possible in Tu~ita heaven (Schlingloff 1988: 4, ch. I, fig.l,l; Schlingloff 1999: pI.23). 145 Caves 16 is apparently slightly earlier since it is thought that this cave stands at the beginning of the fifth century building phase (Spink 1974: 757; ScWingloff 1988: 14; Spink 1991: 71-99).

146 Yazdani 1930-55: IT, pI.XIX; Okada 1996: 150-151. Line drawings: Schlingloff 1988: ch.2, fig.l,1 (cave 16) and ch.2, fig.2,l (cave 2). 147 See ScWingloff 1999: pls.17, 37-38. 148 Schlingloff 1988: 14.

70 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art

The painted lives of Ajll1)!a are the latest extended cycles of the Buddha's life until those preserved in the Western Himalayas. In the sculptural art preserved from the second half of the first millennium the life of the Buddha only occurs in the form of the major events. 149 Eventually, these are canonised as the eight great events, four life events and four great miracles, each ofthem associated with a pilgrimage place. ISO As has been already apparent from the sculptural art discussed so far, the depiction of the events becomes increasingly iconic in character. This trend continues and the narrative element becomes more and more reduced until it is barely recognizable in the art of the Pala and Sena dynasties in north-east India. Klimburg-Salterl51 has tried to bridge the hiatus in Tu~ita depictions with her identification of certain icons as representing the Buddha-to-be in Tu~ita heaven. These depictions have no narrative context and their identification is regionally and temporally confined to the late Buddhist art in North-westIndia (presentday Kashmir, Pakistan and Afghanistan). 152 Also in their iconography these depictions are not directly comparably with the narrative ones. 153 Martin Lerner published a bejewelled and crowned Buddha similar to those in the Hindu Kush that actually occurs in the context of the Buddha's life. 154 Besides this main scene the carving also shows the birth

149 This development is already visible in the art of Nagarjunakol).<;la, when major scenes of the life of the Buddha are collected to be represented on the ayaka-platform frieze facing the entering visitor. The best example for this development is a panel of Site 3 attributed to the late 3rd century (Rosen Stone 1994: figs.158, 209, Dehejia 1997: fig. 157). ISO Early variants ofcanonised events in the Gupta period are discussed in Williams 1975. See also Karetzky 1992: chapter 12. 151 Klimburg-Salter 1988: 200-202.

152 The examples Klimburg-Salter uses are from the Hindu Kush, 7th - 9th centuries, and Greater-Kashmir (the wider region around the Kashmir valley), 8th to 10th century (see Klimburg-Salter 1989: 99-104).

153 In this context the Buddha-to-be wears monastic dress and jewellery including a crown, a bejewelled cape, have'shoulder effulgences' in the shape of a lotus rosette on a crescent, a begging bowl (plUra) and perform vitarkamudra (Klimburg-Salter 1988: 200-202). 154 Lemer 1985: no.26.

71 Christian Luczanits and the Mara's temptation. On the main scene the bejewelled Buddha in monastic dress sits in meditation on a lion throne and teaches (dharmacakramudrii). He is flanked by two four-armed gods or Bodhi­ sattvas, each ofthem with a flask as one attribute and seated with the legs crossed at the ankles on small stools. Other gesturing figures are depicted underneath them while the seven jewels (only three of them preserved) are carved on the lower edge. In the upper part ofthe relief are musicians. Fromthe depiction it remains unclear ifthis scene has to be read narratively or iconic. Both the four-armed attendants and the seven jewels would be unusual in the context of the Buddha's life. I thus tend to read this scene as an iconic one representing the trans-temporal and trans-spatial nature of a fully matured Bodhisattva who displays the life of a Buddha for the benefit of all beings. 155 In Central Asia, too, there are no depictions that can contribute to the development of the Tu~ita episode in the context of the Buddha's life. 156 To see the final development of the story as recounted in the literature and as it may well have once been represented in India as well, one has to look beyond the Indian subcontinent. The stupa ofBarabu<;lur, on the Indonesian island of Java, was begun towards the end of the eighth century and finished in the rust decades of the ninth. At the beginning of the tenth century a new base was added. 15? The huge stupa is also exceptional with regard to the life of the Buddha. There are as many as 120 scenes arranged in chronological succession dedicated to the subject and their literary source is known, the Lalitavistara. Thus, the story ends with the rust preaching at Sarnath. 158 There are no depictions of the teaching events, the miracles or the parinirvii1J-a in Barabudur.

155 A similar concept is used in the Alchi Sumtsek, where the Buddha's life is shown on the dress of Bodhisattva Maitreya (see Luczanits 1999).

156 The most extensive depiciton of the Buddha's life in the cave with the staircase (TreppenhoWe) of Kizil / Quizil begins with Maya's dream (Yaldiz 1987: 79-80, abb.45,1).

15? Dumar~ay 1978: 4-5; de Casparis 1981: 59; Williams 1981: figure 4; Nou, Fr6d6ric 1994: 46. 158 Krom 1927: 227-229.

72 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art

Although the text is followed precisely, there are some peculiarities to the depictions that make the identification of the different scenes difficult and reduce their value as indicators how the events may have been depicted in contemporaneous India. The scenes are crowded with secondary figures not directly related to the event itself. The main figure can only be identified by its position and may be depicted differently in the diverse scenes although it is supposed to represent the same person. The action to be represented is often reduced so much that it cannot be recognized any more. 159 The peculiarities may go back to the production process, since the artists are thought to have worked on the basis of keywords engraved on the stone. 160 The Life of the Buddha is shown in the upper row of the first gallery wallandsurroundsthemonumentinthedirectionofritualcircumambulation (pradakoJi/:za). The first scene shows the Bodhisattva in Tu~ita heaven seated in a palace, bejewelled and with vitarkamudrii (scene 1).161 The gods and goddesses attending him play musical instruments or hold offerings. In the following picture (scene 2) the Bodhisattva announces his rebirth. He is shown as in scene I, but in half proftle on the right side of the panel. Scenes 3 and 4 depict the teaching of the Brahmans and the god's proclamation ofthe Buddhas birth to the Pratyekabuddhas. Scene 5 is dedicated to chapter 4 of the LV, where the Bodhisattva teaches the gods. The following picture shows him bestowing his crown to Maitreya (scene 6).162 The Bodhisattva sits on the central throne and is directed towards the standing Maitreya to his right. His hands are raised as if he has just given his crown to him. The crown, held by Maitreya in both hands, is only preserved in its outlines, but it is clear that it does not differ

159 These peculiarities are even more evident in the depictions of Sudhana's pilgrimage from the Ga.I).<;lavylihasutra (see Fontein 1966: 148). 160 Fontein 1966: 15D-151. For example, a throne is always shown as lion-throne and not only where it is mentioned in the text. 161 See Krom for a detailed description of the different scenes (Krom 1927: I, 101-115, rn, Series La, pls.I-Vll). The numbers given here are taken from his description. Further depictions of all panels: Nou, Frederic 1994: 232-234, Ia1-13. 162 Krom 1927: series La, Plate Ill.

73 Christian Luczanits from those the Bodhisattva and the surrounding gods wear. 163 Again, the attending gods hold different kinds of offerings. Following the consultation of the gods concerning the fonn to be taken for the descent (scene 7) two scenes are dedicated to queen Maya with king Suddhodana (scenes 8 and 9). Two more pictures show the discussion among the gods (scene 10) and the assembly at the time of the Bodhisattva's departure (scene 11). The Bodhisattva, depicted in medi­ tation, then descends in a palace carried by the gods (scene 12).164 Along with the scene of bestowing the crown this scene is the clearest indication that the depictions are in concordance with the description in the LV. In the following scene, the dream of Maya, the Bodhisattva is shown as an elephant with an umbrella floating above him (scene 13). This survey ofthe visual sources yielded comparativelyfew depictions dedicated to the scenes prior to birth with the exception ofMaya's dream. In addition, while Maya's dream is easy to recognize even when the context is lost, the identification of other relevant events is often problematic. Indeed, such scenes can only be identified with certainty once they are part of a chronologically arranged sequence. Further, there appear to be considerable differences from site to site concerning the selection of the events and their rendering. For an interpretation of the overall development of the topic it is crucial to determine if the Tu~ita is actually depicted in Bharhut or not. I have chosen not to accept the customary identification since neither the inscription nor the context ofthe panel give a decisive clue. Nevertheless, even if this scene indeed represents the Bodhisattva in Tu~ita heaven it is

163 Taking the picture at face value, one tends to read the action in the opposite direction: "It seems that the text has not beenfollowed literally, the Bodhisattva does not put the tiara straight on to the head ofMaitreya, and we can not be sure which ofthe two is the Bodhisattva and which Maitreya. One might think that the person seated on the throne here, as elsewhere, must be the Bodhisattva, but the gesture ofthe hands of thisfigure is not that ofsome one who has offered something; thefigure standing is thus evidently the Bodhisattva who has just removed his tiara and is on the point ofgiving it to Maitreya." (Krom 1927: I, 106). This reading is contradicted by the fact that the standing figure, who then would be Sakyamuni, has no u~~f~a. However that may be, the way the scene is represented indicates that the artist did not have a clear idea of the action to be represented and the function of the respective protagonists within it. 164 Krom 1927: Series l.a, pl.v.

74 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art from a tradition that has not yet come down to us in any form. 165 Once Tu~ita can be securely identified, as at Amaravatr, its iconography is fairly stable with a central god surrounded by fellow gods only. While early representations emphasise the veneration later ones show also elements of communication, a development that parallels the texts. With the exception of the extended cycle in the Andhra reliefs and Barabu<;lur, the descent from heaven is not depicted as a separate scene. With the exception of the LV the texts, too, contain no details about the descent. Their focus is either the way the Bodhisattva has to appear in Maya's dream (Nd, MV, MSV, LV), or the form in which the Bodhisattva is to descend (MV, LV). Accordingly the descent is commonly depicted as the dream of Maya and, from the earliest depictions onwards, in fairly consistent iconography.

Interrelation and Transformation Both, textual and visual sources on the events preceding the birth show that they are in essence peripheral to the story of the Buddha's life. Following Bareau the only reason for inventing these stories is the exaltation ofthe historical personage to a divine one. 166 Since the effect is the goal, the range of means that can be used to achieve this effect can be quite varied in the case of a less canonised minor event. Thus, both also show a continuous expansion of the topic over time and considerable variation in detail. In the textual sources the details of the event may be less important than, for example the narrative logic of the text, its doctrinal background and the educational and/or psychological effect that is to be achieved with the story as a whole. The sources themselves make it also clear, that not everything can be changed at free will. In our case, pre- or non­ Buddhist beliefs and memorized verses are the conservative force ham­ pering the development in the case of established events. The donor, his religious advisor and the always anonymous artist, too, may wish to achieve a certain effect based on an overall concept. For

165 Neither could there be a text found that can be directly related to the inscription nor is the Tu~ita known as being inhabited by winged beings.

166 Bareau 1995: 266.

75 Christian Luczanits us today, this desired effect of the narrative is much harder to grasp since commonly only fractions of it are preserved. In addition, from its beginnings Buddhist visual culture is amalgamated with a more mundane vocabulary of fertility, fortune and marvel. The conservative element may here be coined visual vocabulary and visual tradition. Given how little a whole group oftexts says about the events prior to birth, it is surprising that the dream of Maya features so prominently in the earliest Buddhist artthat it can even be used as a marker ofKapilavastu, as in the SanC! depiction. The enormous size ofthe elephant in both early depictions and it's characterisation as royal domestic animal at Bharhut may well indicate that these depictions precede the textual versions that have come down to us. In any case, they appear to be the result of a narrative tradition concerning great births in a royal context. The scarce information in the earlier texts, such as the MASu, also make it unlikely that any Tu~ita depiction preceded the beginning of the Common Era. In fact, the prominence of the depiction of the seven Buddhas of the past on the gateways of SanC! Stiipa 1 makes them, as a group, the most important topic represented and the only one that can be considered programmatic. 167 At Bharhut, too, medallions are dedicated to the seven Buddhas. However one interprets this probably undervalued fact, it is clear that these early depictions, like the MASu, are oriented solely towards the past Buddhas. 168 The MASu, however, only mentions the descent from Tu~ita and does not dwell on the sojourn there. These all are reasons to view the identification of the Tu~ita scene at Bharhut with scepticism. In fact, only in the third century can the Tu~ita heaven episode be clearly identified. There the events are shown in a form that remained peculiar to the region and time, namely third to fourth century Andhra. In Gandhara, in contrast, the last life of the Buddha usually commenced with Maya's dream. The Gandharan convention may also have been instrumental for the Qizil cycle of the Buddha's life. The Tu~ita scene of

167 See Dehejia 1997: 120,29.

168 Since in all cases the group hinted towards consists ofseven, it can be excluded that any of the trees refers to Maitreya, as has been assumed for one separate panel on the south gateway Dehejia 1997: appendix three.

76 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art the Sikri Stiipa remains unique and is also fairly late (circa fifth century). Nevertheless, it is likely that the expansion of the episodes in Tu~ita heaven goes back to developments in Gandharan Buddhism, but these developments were independent of the life of the Buddha. Depictions of the Bodhisattva Maitreya are first found in art ofthe KU~fu:1a realm, either made in Gandhara itself or in Mathura. Indeed, already by the time of Kani~ka (second quarter of the first century CE), Maitreya was deemed important enough to be shown on copper coins. It is not clear if already then Maitreya was thought to be residing in Tu~ita heaven, but this notion certainly soon gained popularity in the region, as numerous legends of visits to Maitreya testify.169 Maitreya and the development of his iconography play an integrative role in the development ofthe life scenes prior to birth the discussion of which goes far beyond the aims of this article. I 15 With regard to the development of the Buddha's life, too, Gandharan art is of pivotal importance. No general program for the Buddha's life events can be extracted from Bharhut or Saiicil70 and if scenes are interlinked they are commonly so on the basis of spatial considerations. Spatially guided arrangements also rule most ofthe narrative paintings of Aj~tiLGandharanart, incontrast, notonlyinventedthe strictchronological sequence for the life of the Buddha,171 but also shows a changed attitude towards the Buddha's life. Rather than focusing on the previous and last lives of the past Buddhas and the veneration of the relic-traces they left, Gandharan art emphasises the development ofthe past individual besides turning it's focus towards the present and the future. Beginning with the first oath in front of the remote predecessor Dlparpkara, the Gandharan life of the Buddha emphasises the spiritual career of a Buddhist aspirant across many lives culminating in the final live of a Buddha. Similar more comprehensive concepts for the Buddha's life are visible in the Nd and the MV. Both commence with the remote past and

169 Most important are those concerning the Maitreya image at Darel, present-day Northern Pakistan, and the five Maitreya(natha) texts. 170 Dehejia 1997: 97-101, 22-30. 171 Taddei 1993; Taddei 1999.

77 Christian Luczanits

the oath in front of Buddha DlpaJ11kara. The Nd then follows successive previous lives of Sakyamuni until, as Vessantara, he becomes reborn in Tu~ita heaven with only one life remaining. With DlpaJ11kara extending the row of Buddhas further into the past, this concept essentially is fully compatible with the older notions. The MV takes a much more innovative approach by connecting the prophesied Bodhisattva with each of the intermediate Buddhas. Each of these Buddhas re-enforces the prophesy and the Bodhisattva is thus linked directly to all his predecessors. Combining this element with the career of a Bodhisattva, the MV represents an early stage of Mahayana ideas. Both texts may well have incorporated the Gandharan emphasis on the spiritual career. AmaravatI and Nagarjunakol).<;la are commonly interpreted in comparison to and in direct continuation ofthe earliest Buddhist art, since in the early Andhra reliefs the Buddha is not yet depicted in human form. 172 However, the narrative cycles in the reliefs ofthe Ik~vaku period (3rd to early fourth century)173 clearly testify the incorporation of some of the developments summarized for Gandharan art. For example, the relevant Andhra reliefs of the extended birth cycle are arranged in chronological sequence, although this is not the only mode scenes of the Buddha's life are depicted there. Among the literary sources only one verse ofthe MV clearly indicates a descent in the form of an elephant as it is found in AmaravatL But the MV, or any other text, does not account for all details represented such as the descent in a palanquin carried by the gods. Thus, the extended birth cycle of AmaravatI is more explicit than any of the preserved literary sources. As with the dream at Bharhut, the cycle likely represents a version ofthe legend that has not come down to us. This version also had a restricted temporal and regional spread and one may therefore speak of a local tradition. This tradition, visually or literary, may in turn have influenced the descent in a heavenly pavilion as it entered the final version of the LV.

172 A notable exception is the detailed study of NagarjunakoQ.~a by Rosen Stonel994. 173 Rosen Stonel994: 12-20.

78 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

To further assess the relationship ofthe Andhra depictions to the MV, one may look at the episode of the visit to the temple immediately after the birth that is characteristic to the Andhra representations. In the MV it is the temple of goddess Abhaya,174 while the Andhra depictions show a god. 175 A similar episode is narrated in the MSV, where the deity is the y ak~a Sakyavardhana. 176 Again, the MV is close in general, but not in detail. Remarkably, the Andhra depictions also do not continue the focus on the past Buddhas that is so characteristic for Sanc! and still omnipresent in Gandharan art. Obviously, with a wide range of Jatakas depicted the exemplary past deeds of the Buddha remain an important matter, but his predecessors apparently were not depicted at all. Depictions of Maitreya in Tu~ita as well as of other Buddha worlds or paradises in Gandharan art177 testify a new focus towards the present and a more or less remote future. His presence in Tu~ita heaven makes Maitreya immediately accessible and rebirth at the time of his last life becomes a remote goal. The elaboration of the Tu~ita events, especially the depiction ofthe sojourn there, likely have to be interpreted in the light of these decisive developments. LaterGandharanreliefs (fourthcenturyandlater) showtheBodhisattva Maitreya teaching in Tu~ita heaven. By that time the Bodhisattva ideal as it is expressed in Mahayana Buddhism was already fully developed. In Ajal).~a, by the end of the 5th century, Sakyamuni, too, is teaching in Tu~ita. The earliest account of the Buddha legend describing the Bodhi­ sattva teaching there is the MSV. Although this links the AjaI).~a depiction and the MSV their relationship is not visible in any further detail. Since long, the Bodhisattva has developed into an ideal to be followed and the Buddha images became the focus of cultic activity. Buddhas and Bodhisattvas are considered to be present at any time and in any world

174 Senart 1882: 1,223 and 11, 26.

175 For example Knox 1992: no.60; Dehejia 1997: fig. 154. 176 Gnoli 1977: I, 48. Schlingloff has identified the scene for the AjaI)ta depictions ofthe Life as well (Schlingloff 1999: 17,13 and 37,13).

177 This is how I interpret depictions such as the Moharnmad Nari stele that are often associated with the Great Miracle at Sravastl.

79 Christian Luczanits system and only display their exemplary deeds for the benefit of all beings. The artistic programs ofCave 2 and 16 at AjaI).ta demonstrate this relationship visually. Both life representations are found on the side walls of the respective caves and in their entirety once covered approximately a third of the wall. 178 Nevertheless, they are just one narrative among many other Jatakas. The main topics found on the main wall and in the Cella area are miraculous events, repeated Buddha representations, Bodhisattva icons and the cult images in the cellas themselves. Thus, rather than being a main topic, at AjaI).ta the life ofthe Buddha has become one among a number ofedifying birth stories. Both the Bodhisattva ideal and the deification of the Buddha opened up new ways to final release that make the exemplary life of the Buddha less and less relevant. As a result the life gets reduced to main events that become increasingly iconic in character. With the establishment of Buddhism in new areas, such as Java and later Tibet, the life ofthe Buddha attains new importance again. The extensive depiction at Barabu<;lur, based on an extended textual source that has just reached its [mal stage, is evidence of this trend. Only here, the transmission of the crown to Maitreya, his successor in Tu~ita heaven, is actually depicted. This curious and unique episode can be seen as a continuation and solidification of the interconnection of future Buddhas with those of the past, as it is already expressed in the MV. The bestowing of a crown, is a clear reference to both, the anointment of a prince as successor and the Tantric initiation rite. In both rituals, the simple diadem ofearlier periods was replaced by the more complex crown. One may thus conclude, that the development of the Tu~ita episodes reflect all major steps in the development ofBuddhism itself, from the oldest teachings to the emerging way of the mantra (Mantrayana) or diamond vehicle (Vajrayana).

178 See Schlingloff 1999: pls.13-20, 28-38.

80 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

Bibliography: Alam, Humera. 1994. "The Sikri Stupa in Lahore Museum. A fresh study of its reliefs." Lahore Museum Bulletin VII, no. 1-2: 103-118, pIs. 1-VII. ARASI = Annual Reports of the Archaeological Survey of India. Bareau, Andre. 1962. "La legende de la jeunesse du Buddha dans les Vinayapitaka anciens." Oriens Extremus 9, no. 1: 6-33. -.1974. "Lajeunesse du Buddha dans les Siitrapifaka et les Vinayapifaka anciens." Bulletin de l'Ecole fran~aise d'Extreme-Orient 61: 199­ 274. -. 1995. Recherches sur la biographie du Buddha dans les Siitrapifaka et les VinayapifakaanciensIII.Articlescomplementaires,Monographies, no 178. Paris: Presses de l'Ecole fran9aise d'Extreme-Orient. Barua, Benimadhab. 1979. Barhut. Patna: Indological Book Corpo­ ration. Bays, Gwendolyn. 1983. The Voice of the Buddha, The Beauty of Compassion. 2 vols, Lalitavistara Siitra, 'Phags-pa rgya-cher-rol-pa zhes-bya-ba theg-pa chen-po'i mdo. Oakland: Dharma Publ. Beal, Samuel. 1875. The romantic legend ofSiikya Buddha. A translation ofthe Chinese version ofthe Abhini~krama1Jasiitra. London: Trtibner (reprint New Delhi 1985: Motilal Barnassidass). -.1883. The Fo-Sho-Hing-Tsan-King Buddhacarita: A Life ofBuddha by Asvagho~a Bodhisattva translated from Sanskrit into Chinese by Dharmarak~a, AD 420. Oxford: Clarendon Press (Sacred Books of the East, Vol. 19). Bechert, Heinz (ed.). 1994. Sanskrit-Worterbuch der buddhistischen Texte aus den Turfan-Funden und der kanonischen Literatur der Sarviistiviida-Schule, Band I, Vokale. Gottingen: Vandenhoeck & Ruprecht. Burgess, lames. 2005. Notes on the Amaravati Stupa. Delhi: Bharatiya Kala Prakashan. -. 1886. The Buddhist Stupas of Amaravati and Jaggayyapeta in the Krishna District, Madras Presidency, surveyed in 1882. Reprint Varanasi: Indological Book House 1970.

81 Christian Luczanits

Chalmers, Lord. 1927. Further Dialogues ofthe Buddha translatedfrom the Pali of the Majjhima Nikaya. 2 vols. London: Humphrey Milford. Chalmers, Robert. 1899. The Majjhima-Nikaya. 4 vols. London: Pali Text Society, (reprint 1977). Cowell, Edward B. 1893. The Buddha-Karita or Live of Buddha by Asvaghosha, Indian poet of the early second century after Christ. Repring Amsterdam: Oriental Press 1970. CPD. A Critical paliDictionary. Edited by V. Trenckner, Dines Anderson and Helmer Smith. Vol. 1. Copenhagen, 1924. Csoma de Koros, Alexander. 1839. "Notices on the Life of Shakya extracted from the Tibetan Authorities." Tibetan Studies (reprinted 1986): 229-263. Cunningham, Sir Alexander. 1879. The Stupa of Bharhut. London: H. AlIen and Co. (Reprint Delhi, Munishiram Manoharlal, 1998). Czuma, Stanislaw J. 1985. Kushan Sculpture: Images from Early India. Cleveland: The Cleveland Museum of Art. Dar, Saifur Rahman. 1999/2000 "The Sikri sculptures: prolegomena on an exceptional, but unstudied, collection of Gandharan art in the Lahore Museum." Silk RoadArtandArchaeology 6, Papers in honour of Francine Tissot: 19-43. de Casparis, J. G. 1981. "The Dual Nature of Barabu<;lur." Barabu{iur. History and Significance ofa Buddhist Monument: 47-83. de Jong, J.W. 1954. "L'episode d'Asita dans le La1itavistara." In: Schubert, Johannes; Schneider, Ulrich. Asiatica. Festschrift Friedrich Weller. Leipzig: Otto Harrassowitz: 312-325. -.1990. "Buddhacarita VIII.21." Indo-Tibetan Studies. Papers in honour and appreciation ofProfessor David L. Snellgrove's contribution to Indo-Tibetan Studies, Buddhica Britannica, Series Continua 11: 145­ 146. Dehejia, Vidya. 1997. Discourse in Early BuddhistArt: Visual Narratives ofIndia. New Delhi: Munshiram Manoharlal. Dumar~ay, Jacques. 1978. Borobudur. KualaLumpur: Oxford University Press. Dutoit, Ju1ius. 1921. Jatakam. Vol. 7. Leipzig.

82 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art

Edgerton, Franklin. 1953. Buddhist Hybrid Sanskrit Grammar and Dictionary. 2 vols. New Haven: Yale University Press. Faccenna, Domenico. 1964. Sculptures from the Sacred Area ofButkara I (Swat Pakistan). 3 vols. Vol. 2 and 3. Rome: Istituto Poligraphico dello Stato. Fausb011, V. 1877. The Jiitaka together with its commentary being tales ofthe anterior births ofthe Buddha. Vol. 1. Oxford: Pali Text Society (reprinted 1990). Fergusson, James. 1873. Tree and Serpent Worship or Illustrations of Mythology and Art in India in the First and Fourth Centuries after Christ from the Sculptures of the Buddhist Topes of Sanchi and Amaravati. New Delhi: Asian Educational Services (reprinted 2004). Fontein, Jan. 1966. The Pilgrimage ofSudhana. A Study of Ga1J~avyuha Illustrations in China, Japan and Java. The Hague-Paris. Foucaux, E. 1884-92. Le LaUta Vistara. 2 vols. Paris. Foucher, M. A. 1903. "Les Bas-Reliefs du Stiipa de Sikri (Gandhara)." Journal Asiatique Sept. /Oct.: 185-330. -.1905-51. L'Art Greco-Bouddhique du Gandhiira. 3 vols. Paris. -. 1949. La Vie au Bouddha d'apres les textes et les monuments de l'Inde. Paris. Fukita, Takamichi. 2003. The Mahiivadiinasutra. A new edition based on manuscripts discovered in northern Turkestan. Gottingen: Vandenhoeck & Ruprecht. Gies, Jacques, and Monique Cohen (eds.). 1996. Serinde, Terre de Bouddha: Dix siecles d'art sur la Route de la Soie. Paris: Editions de la Reunion des musees nationaux. Gnoli, Raniero. 1977. The Gilgit Manuscripts ofthe Sanghabhedavastu. 2 vols. Vol. XLIX,l, Serie Orientale Roma. Roma: IsMEO. Gronbold, Gtinter. 1984. "Die Mythologie des indischen Buddhismus." In Worterbuch der Mythologie, edited by Hans Wilhelm Haussig, 287-507. Stuttgart: Emst Klett Verlag. Harle, J. C. The Art and Architecture ofthe Indian Subcontinent. 1994. New Haven-London (2nd ed). Homer, I.B. 1967-76. The Collection of Middle Lenght Sayings (Majjhimanikiiya). 3 vols. London: Luzac & Company.

83 Christian Luczanits

-. 1975. The Minor anthologies ofthe Pala Canon, part Ill. Chronicle of the Buddhas (BuddhavarJ"lsa) and Basket ofConduct (Cariyiipi!aka). London: The Pali Text Society. Ingholt, Harald. 1957. Gandhiiran Art in Pakistan. New York: Pantheon Books. Jamspal, L. 1999. Saundaranand Mahiikiivya ofAciirya Asvagho~a with Tibetan and Hindi Translations. Sarnath, Varanasi: Central Institute of Higher Tibetan Studies. Janert, Klaus Ludwig. 1977. "KalsI-Elefant und EmpHingnisvision der BuddhamutterMaya." Nachrichten derAkademie der Wissenschaften in Gottingen, Philologisch-Historische Klasse 3: 65-76. Jayawickrama, N.A. The Story ofGotama Buddha. The Nidiina-kathii of the liitaka!!hakathii. Oxford: The Pa1i Text Society, 1990. Jayawickrama, N.A. 1995. BuddhavarJ"lsa and Cariyiipi!aka, Pali Text Society, Text Series no. 166. Oxford: The Pali Text Society. Johnston, E.H. 1928. The Saundarananda ofAsvagho~a. Critically edited and translated with notes. Delhi: Motilal Banarsidass (reprinted 1975). -. 1936. Asvagho~a's Buddhacarita or Acts of the Buddha. Complete Sanskrit text with English translation. Delhi: Motilal Banarsidass (reprinted 1978). Jones, J. J. 1949-1956. The Mahiivastu. 3 vols, Sacred Books of the Buddhists (1949, vol. XVI; 1952, vol. XVII; 1956, vol. XIX). London: Luzac & Company. Karetzky, Patricia Eichenbaum. 1992. The Life of the Buddha. Ancient Scriptural and Pictorial Traditions. Lanham, New York: University Press of America. Khosla, Sarla. 1991. Lalitavistara and the Evolution of the Buddha Legend. Delhi: Galaxy Publications. Klimburg-Salter, Deborah E. 1988. "The Tucci Archives Preliminary Study, 2: The Life of the Buddha in Western Himalayan Monastic Art and Its Indian Origins: Act One." East and West 38, no. 1-4: 189­ 214. -. 1989. The Kingdom ofBiimiyiin: BuddhistArtand Culture in the Hindu Kush. Napels: Istituto Universitario Orientale. Dipartimento di Studi Asiatici.

84 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

Kloetzli, W. Randolph. 1983. Buddhist Cosmology. Science and Theology in the Images ofMotion and Light. Delhi: Motilal Banarsidass. Knox, Robert. 1992. Amaravati, BuddhistSculpturefrom the Great Stiipa. London: British Museum Press. Koppen, C. F. 1906. Die Religion des Buddha und ihre Entstehung (1. Band, 2. Auflage). Berlin. Krom, N. J. 1925. The Life of the Buddha on the Stupa of Barabufj.ur according to the Lalitavistara-Text. Delhi: Bharatiya Publishing House (reprinted 1974). -.1927. Barabudur, archeological description. 5 vols. New Delhi: Gyan Publishing House (reprinted 1986). Kurita, Isao. 1988. Gandhiiran Art 1. The Buddha's life story, Ancient Buddhist art Series. Tokyo: Nigensha. Lamotte, Etienne. 1966-80. Le Traite de la Grande Vertu de Sagasse de Niigiirjuna (Mahiiprajfi.iipiiramitiisiistra). 5 vols. Louvain-Ia-Neuve: Universite de Louvain, Institut Orientaliste. -. 1988. History ofIndian Buddhism from the Origins to the Saka Era. Louvain-Ia-Neuve. Lefmann, Salomon. 1874. Lalita Vistara, Erziihlung von dem Leben und der Lehre des Sakya SirJ1ha, Erstes Buch: Vorleben im Tu~itahimmel. Berlin. 1902 / 08. Lalita Vistara, Leben und Lehre des Siikya-Buddha, Textausgabe mit Varianten-, Metren- und Worterverzeichnis. 2 vols. Halle. Lemer, Martin. 1985. The Flame and the Lotus: Indian and Southeast Asian Artfrom the Kronos Collections. New York: The Metropolitan Museum of Art. Leumann, Emst, and Shindo Shiraishi. 1957. Mahiivastu, ein buddhistischer Sanskrittext, Heft Il, Reprintfrom Proceedings ofthe Faculty ofLiberal Arts and Education, no. 2: Yamanashi University Japan. -. 1962. Mahiivastu, ein buddhistischer Sanskrittext, Heft Ill, Reprint from Proceedings ofthe Faculty ofLiberal Arts and Education, no. 3: Yamanashi University Japan.

85 Christian Luczanits

Lienhard, Siegfried. 1984. A History ofClassical Poetry. Sanskrit - Pali - Prakrit, A History ofIndian Literature, Vol. Ill, 1. Wiesbaden: Otto Harrasowitz. Lobo, Wibke. 1987. "Der Bodhisattva mit dem FUischchen: Siddhartha oder Maitreya?" In: Falk, Harry (ed.). Hinduismus und Buddhismus. Festschrift jar Ulrich Schneider. Freiburg: Hedwig Falk: 208-219, abb.7-13. -. 1991. ''The Bodhisattva with the Flask: Siddhartha or Maitreya?" In: Gouriswar Bhattacharya (ed.). Ak~ayanfvf: Essays presented to Dr. Debala Mitra. Delhi: Sri Satguru: 95-104. Lommel, Andreas. 1974. Kunst des Buddhismus aus der Sammlung des staatlichen Museums jar Volkerkunde in Miinchen. ZUrich: Atlantis. Longhurst, A.H. 1935. ARASI. -.1938. The BuddhistAntiquitiesofNiigiirjunak01:uf.a, Madras Presidency. Vol. 54, Memoirs of the Archaeological Survey of India. Delhi: Manager of Publications. Luczanits, Christian. 1999. "The Life of the Buddha in the Sumtsek." Orientations 30 (1): 30-39. -.2005. "The BodhisattvaWith the Flask in Gandharan Narrative Scenes" East and West 55, no. 1-4, dedicated to Maurizio Taddei: 163-88. LUders, Heinrich. 1941. Biirhut und die buddhistische Literatur, Abhandlungen jar die Kunde des Morgenlandes, Band 26, Nr. 3. Leipzig: Deutsche MorgenHindische Gesellschaft (Reprint Nendeln, Kraus Reprint, 1966). - (ed.). 1963 Bharhut Inscriptions, Corpus Inscriptionum Indicarum. New Delhi: Archaeological Survey of India (Reprint 1998). Mallebrein, Cornelia. 1984. Skulpturen aus Indien. Bedeutung und Form. MUnchen. Marcus, Margaret F. 1967. "Sculptures from Bihar and Bengal." The Bulletin ofthe Cleveland Museum ofArt LIV, no. 0: 240-262. Marshall, John. 1916. "Excavations at Taxila." ASIAnnual Report 1912- 1913: 1-52. 1960. The Buddhist Art ofGandhiira, Memoirs ofthe Department of Archaeology in Pakistan, Vol. 1. London: Cambridge University Press. Marshall, Sir John; Foucher, Alfred. 1940. The Monuments ofSiifichf. 3 vols. Calcutta: Govt. of India Press.

86 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

Meunie, Jacques. 1942. Shotorak, Memoires de la Delegation Archeologique Franfaise en Afghanistan X. Paris. Mitra, Rajendra Lal. 1998. The Lalita-Vistara. Memoirs ofthe Early Life ofSakya Sinha (Chs. 1-15). Delhi: Sri Satguru Publications. -. 1877. Lalitavistara, Bibliotheca Indica. Calcutta. Nfu).amoli, Bhikkhu; Bodhi, Bhikkhu. 1995. The Middle Lenghts Discourses of the Buddha. A New Translation of the Majjhima Nikiiya. Kandy, Sri Lanka: Buddhist Publication Society. Norman, K.R. 1983. Piili Literature, including the canonical literature in Prakrit and Sanskrit ofall the Hfnayiina schools ofBuddhism. Edited by Jan Gonda, A History of Indian Literature, Vol. II, Fasc. 2. Wiesbaden: Otto Harrasowitz. Nou, Jean-Louis; Frederic, Louis. 1994. Borobudur. New York - London - Paris: Abbeville Press Publishers. Okada, Amina. 1996. Ajanta. New Delhi: Brijbasi Printers. Oldenberg, Hermann. 1921. Buddha, sein Leben, seine Lehre, seine Gemeinde. Berlin. -.1967. "Studien zum Mahavastu." Janert, Klaus Ludwig (ed.). Hermann Oldenberg. Kleine Schriften. Wiesbaden: Franz Steiner: 1037-1068. Parimoo, Ratan. 1982. Life ofBuddha in Indian Sculptures (Ashta-Maha­ Pratiharya). An Iconological Analysis. New Delhi: Kanak Publications. Pischel, Richard. 1910. Leben und Lehre des Buddha. Leipzig. PW. 1855-1875. Sanskrit-Worterbuch. Edited by Otto Bohtlingk and Rudolph Roth. Reprint Tokyo: Meicho-Fukyii-Kai,1976). Q. 1955-1958. "The Tibetan Tripitaka, Peking Edition." edited by Daisetz T. Suzuki. Tokyo-Kyoto: Tibetan Tripitaka Research Institute. Rahula, Bhikkhu Telwatte. 1978. A Critical Study of the Mahiivastu. Delhi: Motilal Banarsidass Publ.. Rao, Manjushree. 1994. Sanchi Sculptures: An Aesthetic and Cultural Study. Delhi: Akai Book Cooperation. Rao, P.R. Ramachandra. 1956. The Art ofNiigiirjuIJik01}rj.a. Madras. Rhys Davids, T.W. 1998. Buddhist Birth-Stories, Jataka tales. Culcutta: Srishti. Rhys Davids, T.W.; Carpenter, J. Estlin. 1903. The Dfgha Nikiiya. 3 vols. London: The Pali Text Society (reprinted 1995).

87 Christian Luczanits

Rhys Davids, T.W.; Rhys Davids, c.A.P. 1910. Dialogues ofthe Buddha. 4th ed. 3 vols, Sacred Books ofthe Buddhists, Vols. II-N. London: The Pali Text Society (reprinted 1959). Rockhill, W. W. 1884. The Life ofthe Buddha and the Early History of his Order. San Francisco: Chinese Materials Center (reprinted 1976). Rosen Stone, Elizabeth. 1994. The Buddhist Art of Nagarjunako1}{ia. Delhi: Motilal Banarsidass (Buddhist Tradition Series, vol. 25). Scherman, L. 1928. "Die altesten Buddhadarstellungen des Mtinchener Museums fUr Volkerkunde I." Miinchner Jahrbuch der bildenden Kunst V: 274-290. Schiefner, Anton. 1849. Eine Tibetische LebensbeschreibungSakyamuni's des Begriinders des Buddhathums. St. Petersburg. Schlingloff, Dieter. 1988. Studies in the Ajanta paintings: identifications and interpretations. Delhi: Ajanta Publications. -.1999. Guide to the Ajanta Paintings. Vol. 1 Narrative Wall Paintings. Delhi: Munshiram Manoharlal. -. 2000. Ajanta - Handbuch der Malereien / Handbook ofPainting 1: Erziihlende Wandmalereien / Narrative Wall-paintings. 3 vols. Wiesbaden: Harrassowitz Verlag. Schmidt, Richard. 1923. Buddha's Leben. Asvaghosa's Buddhacaritam. Ein altindisches Hedengedicht des 1. Jahrhunderts n. Chr. Hannover: Orient-Buchhandlung Heinz Lafaire. Schopen, Gregory. 2004. "Art, Beauty, and the Business of Running a Buddhist Monastery in Early Northwest India." In: Buddhist Monks and Business Matters. Still More Papers on Monastic Buddhism in India. Honolulu: University of Hawai'i Press: 19--44. -. 2005. "On Sending the Monks Back to their Books. Cult and Conservatism in Early Mahayana Buddhism." In: Figments and Fragments ofMahayana Buddhism in India. More Collected Papers,. Honolulu: University of Hawai'i Press: 108-153. Schotsmann, Irma. 1995. A§vagho~a's Buddhacarita. The Life of the Buddha, Bibliotheca Indo-Tibetica Series XXXI. Sarnath, Varanasi: Central Institute of Higher Tibetan Studies. Schubring, Walther. 1954. "Zum Lalitavistara." In: Schubert, Johannes; Schneider, Ulrich (ed.). Asiatica, Festschrift Friedrich Weller. 610­ 655. Leipzig: Otto Harrassowitz.

88 Prior to Birth - The Tu~ita episodes in Indian Buddhist literature and art

Senart, E. 1882. Le Mahiivastu. 3 vols. Paris. Shanna, R.c. 1994. Bharhut Sculpture. New Delhi: Abhinav. Sivaramamurti, Calambur. 1942. "Amaravati Sculptures in the Madras Government Museum." Bulletin ofthe Madras Government Museum New Series, General Section, Vol. IV. -.1974. The Art ofIndia. New York: Harry N. Abrams. Snellgrove, David L. (ed.). 1978. The Image ofthe Buddha. Paris: Serindia /UNESCO. Spink, WaIter. 1974. "The Splendours of Indra's Crown: A Study of Mahayana Developments at AjaI).ta." Journal ofthe Royal Society of Arts 122: 743-767. -. 1991. "The Vakataka's Flowering and Fall." In: Parimoo, Ratan; Kannal, Deepak; Panikkar, Shivaji; Poduval, Jayaram; Shanna, Indramohan (ed.). The Art ofAjanta, New Perspectives. New Delhi: Books & Books: 71-99. Stede, W., ed. Sumangala-viliisinf, Buddhaghosa's Commentary on the Dzgha-Nikiiya. 3 vols. Vol. 2, 1971. Taddei, Maurizio. 1993. Arte narrativa tra India e mondo ellenistico. Rome. -.1999. "Oral Narrative, Visual Narrative, Literary Narrative in Ancient Buddhist India." India, Tibet, China. Genesis and Aspects of Traditional Narrative: 71-85. -.1995. "Was bedeutete der Buddhismus fUr die frtihe indische Kunst?" In: Klimburg-Salter, Deborah(ed.). Buddhain Indien. Diefriihindische Skulptur von Konig Asoka bis zur Guptazeit (Exhibition catalogue Kunsthistorisches Museum Wien). Vienna: Skira: 41-49. Tamura Jitsuzo; Imanishi Shunjii; Sato Osa. 1966-68. Gotai shinbunkan yakukai = Wu ti qing weng jian (Mahiivyutpatti). 2 vols. Kyoto: Kyoto Daigaku Bungakubu Nairiku Ajia Kenkyii. Vaidya, P. L. 1958. Lalita-Vistara, Buddhist Sanskrit Texts No. 1. Darbhanga: The Mithila Institute. Vogel, Claus. 1970. The Teachings ofthe six Heretics according to the Pravrajyiivastu ofthe Tibetan Miilasarviistiviida Vinaya edited and rendered into English, with an appendix containing an English translation ofthe pertinent sections in the Chinese Mulasarviistiviida Vinaya, AbhandlungenfUr die Kunde des Morgenlandes, Bd. XXXIX. Wiesbaden.

89 Christian Luczanits van Hiniiber, Oskar. 1994. Untersuchungen zur Mundlichkeit fruher mittelindischer Texte der Buddhisten (Untersuchungen zur Sprachgeschichte und Handschriftenkunde des Piili Ill), Akademie der Wissenschaften undLiteratur, Maim. Abhandlungen der Geistes­ und Sozialwissenschaftlichen Klasse 5. Stuttgart: Franz Steiner. -. 1998. Entstehung und Aujbau der Jiitaka-Sammlung. Studien zur Literaturdes Theraviida Buddhismus I, Akademie der Wissenschaften und Literatur, Mainz. Abhandlungen der Geistes- und Sozialwissenschaftlichen Klasse 7. Stuttgart: Franz Steiner Verlag. Waldschmidt, Emst (ed.). 1953. Das Mahiivadiinasutra, ein kanonischer Text uber die sieben letzten Buddhas, Abhandlungen der deutschen Akademie der Wissenschaften. Berlin: Akademie-Verlag. Walshe, Maurice. 1996. The Long Discourses ofthe Buddha. A translation of the Dfgha Nikiiya. Kandy, Sri Lanka: Buddhist Publication Society. Weller, Friedrich. 1926-28. Das Leben des Buddha von Asvagho~a. Tibetisch und Deutsch. 2 vols. Leipzig: Eduard Pfeiffer. -. 1939. "Schauplatz und Handlung im Buddhacarita." Zeitschrift der Deutschen Morgenliindischen Gesellschaft 93 (= N.F. 18): 306­ 338. -. 1987. "Zum Lalita Vistara I. Uber die Prosa des Lal. Vist." In Kleine Schriften, edited by Wilhelm Rau, 457-510. Stuttgart. 1953. Zwei zentralasiatische Fragmente des Buddhacarita, Abhandlungen der siichsischen Akademie der Wissenschaften zu Leipzig. Philologisch-historische Klasse, Band 46, Heft 4. Berlin: Akademie Verlag; Reprint Rau, Wilhelm ed. (1987) Friedrich Weller. Kleine Schriften. 2 vols. Wiesbaden, Franz Steiner: 343-369. Williams, Joanna Gottfried. 1975. "Sarnath Gupta steles of the Buddha's life." Ars Orientalis 10: 171-192. -. 1981. "The Date of Barabu<;lur in Relation to Other Central Javanese Monuments." In Baraburj.ur, History and Significance ofa Buddhist Monument, 25-45. Berkeley. Windisch, Emst. 1908. Buddha's Geburt und die Lehre von der Seelenwanderung. Leipzig: B.G. Teubner (Abhandlungen der philologisch-historischen Klasse derkonigl. Sachsischen Gesellschaft der Wissenschaften).

90 Prior to Birth - The Tu#ta episodes in Indian Buddhist literature and art

-.1909. Die Komposition des Mahiivastu. Ein Beitrag zur Quellenkunde des Buddhismus, Abhandlungen der philologisch-historischen Klasse derkonig!. sachsischen Gesellschaftder Wissenschaften XIV. Leipzig: B.G. Teubner. Wintemitz, Maurice. 1983. A History ofIndian Literature. Translated by V. Srinivasa Sarma. 2 vols. Vol. 2, Buddhist and Jaina Literature. Delhi: Motila1 Banarsidass (reprinted 1999). Yaldiz, Marianne. 1987. Archaologie und Kunstgeschichte Chinesisch­ Zentralasiens (Xinjiang). Leiden: Brill. Yazdani, G. 193Q-55.Ajanta. The colourandmonochrome reproductions ofthe Ajantafrescoes based on photography, Vo!. I, Il, IV. London.

91 Plates and Figures

Luczanits 1

"The descent of the Bhagavat", Bharhut, vedika pillar; Indian Museum. Calcutta, Acc. No. 93; after Coomaraswamy 1956: fig.61. Plates and Figures

Luczanits 2 a & b

Tu~ita, descent and dream, Amaravati, vedika railing; Government Museum Chennai; drawing after Fergusson 1873: pl.lxxiv; image after Sivaravamurti 1974: fig.416.

388

Plates and Figures

Descent from Tu~ita heaven, Amaravati, vedika pillar; Government Museum Chennai; photo C. Luczanits 2007 D5035.

390 Plates and Figures

Luczanits 4

The Bodhisattva in Tu~ita heaven, Ajal!!i'i, Cave 2; after Yazdani 1930-55: n, pl.xix. The Birth of the Buddha

Proceedings of the Seminar Held in Lumbini, Nepal, October 2004

Edited by

Christoph Cueppers, Max Deeg and Hubert Durt

Lumbini International Research Institute Lumbini 2010 Contents

Introduction 1 MaxDeeg

Buddha's Birth and Reassessment of the Archaeological Evidence 19 Giovanni Verardi

Prior to Birth The Tu~ita episodes in Indian Buddhist literature and art 41 Christian Luczanits

Why is the Buddha Riding on an Elephant? The Bodhisattva's Conception and the Change of Motive 93 MaxDeeg

On Some Late Records About the Buddha's Birth and Genealogy 129 Michael Hahn

The Deva with the Swaddling Cloth - On the Western Origins of Gandharan Birth Iconography and Their Implications for the Textual and Art History of the Buddhist Saviour's Nativity 157 Martina Stoye

The Lady under the tree A visual pattern from Maya to the Tara and Avalokitesvara 193 Claudine Bautze-Picron

isionary Consecration: A Meditative Reenactment of the Buddha's Birth 239 obuyoshi Yamabe Lumbini International Research Institute PO.Box 39 Bhairahawa, Dist. Rupandehi NEPAL e-mail: [email protected] © Lumbini International Research Institute

Cover: Nativity of the Buddha, 9th century, stone, h: 84 cm Courtesy of the National Museum, Kathmandu

All rights reserved. Apartfrom anyfair dealing for the purpose ofprivate study, research, criticism or review, no part ofthis book may be reproduced in any form, by print, photocopy, microfilm, scanner or any other means without prior written permission. Enquiries should be made to the publisher.

ISBN 978-9937-553-03-2

Fir t published 2010 Printed in Nepal, Dongol Printers, Kathmandu The Birth of the Buddha in the Chinese Anthologies of the Early Sixth Century 277 Hubert Durt

The Birth of the Buddha and Related Episodes as Represented in Chinese Art 305 Nicoletta Celli

The Birth of the Buddha in Korean Buddhism: Infant Buddha Images and the Ritual Bathing 321 Juhyung Rhi

LumbinI: liturgy and devotion 345 Peter Skilling

The Narrative of the Birth of the Buddha as Told by Bskal-bzang Chos-kyi Rgya-mtsho (15th Century) 355 Franz-Karl Erhard

Plates and Figures 377